ImpactMojoEducation Policy 101www.impactmojo.in
ImpactMojo 101 Series · Free Forever
Education
Policy
101
From enrolment to learning: how South Asian states decide who learns what, who teaches, who pays and how anyone knows whether it worked
100 SlidesSouth Asia FocusFree ForeverRTE to NEP 2020
CC BY-NC-ND 4.0 · ImpactMojo 101 Series01www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Agenda
What we cover
01
Why education policy matters
Slides 3–9
02
The constitution and the RTE Act
Slides 10–18
03
From Kothari to NEP 2020
Slides 19–27
04
Access versus learning: the data
Slides 28–36
05
Paying for school
Slides 37–44
06
Teachers
Slides 45–53
07
Private schools and section 12(1)(c)
Slides 54–61
08
Accountability and assessment
Slides 62–69
09
What works: the evidence
Slides 70–78
10
Neighbours and higher education
Slides 79–87
11
A practitioner's toolkit
Slides 88–94
12
Open debates and where next
Slides 95–99
CC BY-NC-ND 4.0 · ImpactMojo 101 Series02www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
01
Section One
Why education policy matters
CC BY-NC-ND 4.0 · ImpactMojo 101 Series03www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Starting point
What education policy actually decides
Education policy is the set of public choices that decide who goes to school, for how long, what they are taught, who teaches them, who pays and how results are checked. In India those choices are made by Parliament and 36 state and union territory governments, by courts interpreting the constitution, by district officials who post teachers, and by head teachers who decide what happens on a Monday morning.
Rules and rights
The constitution, the Right of Children to Free and Compulsory Education Act 2009, state rules, board regulations and court orders. They set entitlements: a seat in a neighbourhood school, a free midday meal, a qualified teacher.
Money and people
Budgets, centrally sponsored schemes, teacher recruitment and training, textbooks and assessment systems. They decide whether the entitlement turns into a classroom where a child learns to read.
This course follows both halves: what the law promises and what the data show is delivered, with South Asian evidence first.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series04www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Returns to schooling
Each year of school pays, and the evidence is global
Psacharopoulos and Patrinos (2018) pooled 1,120 estimates of the return to education from 139 countries between 1950 and 2014. A private return is the extra earnings a person gains from one more year of schooling, net of what it cost them. Their headline: the average private return is about 9% a year. Social returns, which count public spending as a cost, remain high, and women see higher average returns than men.
9%
average private return to one more year of schooling, worldwide
Psacharopoulos and Patrinos (2018), Education Economics 26(5)
1,120
estimates pooled, from 139 countries, 1950-2014
Psacharopoulos and Patrinos (2018)
Women
have higher average returns to schooling than men
Psacharopoulos and Patrinos (2018)
Returns are an average across years of schooling. They say little about whether a particular year taught anything, which is the next slide's problem.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series05www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
The learning crisis
Schooling is not the same as learning
The World Bank's World Development Report 2018 was the first in the series devoted entirely to education. Its central message is printed in one line: schooling is not the same as learning. Enrolment had risen fast in low and middle income countries, while what children could do after years in class had not kept pace.
In rural India in 2016, only half of grade 5 students could fluently read text at the level of the grade 2 curriculum.
World Bank, World Development Report 2018, Overview, citing ASER Centre (2017)
The same report notes that in rural India just under three-quarters of grade 3 students could not solve a two-digit subtraction such as 46 minus 17, and by grade 5 half still could not.
It calls the learning crisis a moral crisis, and shows that it worsens inequality: it hobbles the disadvantaged youth who most need what a good education can offer.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series06www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Flat learning profiles
Gaps that start early grow wider every year
Learning profile
The relationship between the grade a child is enrolled in and what the child can actually do. A flat profile means each extra year in school adds little skill.
WDR 2018 gives two Indian examples. In Andhra Pradesh in 2010, low-performing grade 5 students were no more likely to answer a grade 1 question correctly than grade 2 students. In New Delhi in 2015, the average grade 6 student performed at a grade 3 level in mathematics, and even by grade 9 had reached less than a grade 5 level, with the gap between stronger and weaker students growing over time.
Why it happens
Curricula move at the pace of the textbook. A child who falls behind in grade 2 faces grade 3 material she cannot follow, and the teacher has no time to go back.
Why it matters for policy
Adding years of school without fixing the profile adds cost and little learning. Section 9 shows the interventions that bend the profile upward.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series07www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
The case for the state
Why families and markets alone under-provide education
Economists give several reasons why governments fund and regulate schooling instead of leaving it to private choice. None of them requires the state to run every school, but each explains why public money and public rules are involved.
ReasonWhat it meansSouth Asian example
Credit constraintsPoor families cannot borrow against a child's future earningsChildren leave school to earn; PLFS 2023-24 data cited by PRS show 44% of out-of-school 14-18 year olds left to supplement household income
Information gapsParents cannot easily see what a school teachesReport cards in Pakistan changed fees and test scores (Section 8)
SpilloversAn educated neighbour, voter or mother benefits othersBangladesh's girls' stipend delayed marriage (Section 9)
Equity and rightsEducation as a constitutional right, owed to every childArticle 21A in India, 25A in Pakistan, 31 in Nepal
Each reason points to a different tool: money for credit constraints, data for information gaps, subsidies for spillovers, law for rights.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series08www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Course map
Four questions this course keeps asking
Every section returns to the same four questions. They are a useful checklist for reading any education scheme, budget line or evaluation you meet in the field.
01
ACCESS: is every child enrolled and attending?
→
02
LEARNING: can they read, write and calculate at grade level?
→
03
EQUITY: who is left behind, by gender, caste, place or disability?
→
04
MONEY: what does it cost and is the money spent?
India has largely answered the first question for children aged 6 to 14: ASER 2024 finds 98.1% of rural children in that age group enrolled.
The second, third and fourth are open, and this course spends most of its time on them: what children learn, who is left behind and what it costs.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series09www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
02
Section Two
The constitution and the RTE Act
CC BY-NC-ND 4.0 · ImpactMojo 101 Series10www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
From directive to right
How education moved from a promise to an enforceable right
When the Constitution came into force in 1950, free and compulsory education for children sat in Part IV, the Directive Principles of State Policy. Article 37 says those principles are fundamental in governance but cannot be enforced by any court. For four decades there was no enforceable right to schooling.
Before 2002
Education was a directive under Article 45. In Mohini Jain (1992) the Supreme Court read a right to education into Article 21; in Unni Krishnan (1993) five judges confined it to free education until a child completes 14.
After 2002 and 2010
The Constitution (Eighty-sixth Amendment) Act 2002 inserted Article 21A as a fundamental right. It came into force on 1 April 2010, together with the RTE Act that gives it content.
The 2002 amendment wrote the judge-made right into the text of Part III and handed its content to Parliament. A parent, a school management committee or a public interest litigant can go to court on the child's behalf.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series11www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
The 86th Amendment
Three articles, three different duties
The Constitution (Eighty-sixth Amendment) Act 2002, which received assent on 12 December 2002, changed three articles at once. Read together they split the duty between the state, for different age groups, and parents.
ArticleText as amendedWho carries the duty
21A (Part III)"The State shall provide free and compulsory education to all children of the age of six to fourteen years in such manner as the State may, by law, determine."The State, enforceable in court
45 (Part IV)"The State shall endeavour to provide early childhood care and education for all children until they complete the age of six years."The State, as a directive
51A(k) (Part IVA)Parents or guardians to "provide opportunities for education to his child or, as the case may be, ward between the age of six and fourteen years"Parents, as a fundamental duty
Early childhood stayed a directive. That is why NEP 2020's push on pre-school rests on policy and schemes, without a justiciable right behind it.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series12www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Who legislates
Education is on the Concurrent List
Entry 25 of List III (the Concurrent List) reads: "Education, including technical education, medical education and universities, subject to the provisions of Entries 63, 64, 65 and 66 of List I; vocational and technical training of labour." It was substituted by the Constitution (Forty-second Amendment) Act 1976, effective 3 January 1977, and State List entry 11 was omitted at the same time.
What the Centre holds
Entry 25 is expressly subject to Union List entries 63 to 66, which keep some subjects with Parliament alone. PRS notes that the Centre is responsible for determining the standards of higher education institutions.
What the states hold
Both can legislate on education generally, but states run most schools, employ most teachers and spend most of the money. PRS describes school education as a shared responsibility, with states responsible for its development.
Source: PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026. Concurrent status explains why a national policy such as NEP 2020 needs each state to adopt it in rules, budgets and recruitment.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series13www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
RTE Act 2009
The core provisions of the RTE Act
The Right of Children to Free and Compulsory Education Act 2009 (No. 35 of 2009) gives Article 21A its content. It came into force on 1 April 2010. Its provisions fall into three groups: the child's entitlement, duties of schools and teachers, and the standards a school must meet.
SectionWhat it says
s3(1)Every child aged six to fourteen has a right to free and compulsory education in a neighbourhood school till completion of elementary education
s12(1)(c)Unaided and specified schools admit at least 25% of the class I (or entry) strength from weaker sections and disadvantaged groups
s13(1)No capitation fee and no screening procedure for the child or parents
s17(1)"No child shall be subjected to physical punishment or mental harassment"
s21Every school except a private unaided one constitutes a School Management Committee of local authority representatives, parents or guardians and teachers
s24(1)Teachers maintain regularity and punctuality and complete the curriculum
s25 and SchedulePupil-teacher ratios and minimum norms for every school
CC BY-NC-ND 4.0 · ImpactMojo 101 Series14www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Rights in practice
What a family can ask for under the RTE Act
Statutes are easier to use when they are turned into plain claims. The list below restates the RTE Act from the point of view of a parent of a child aged six to fourteen. Each claim points to the section that supports it, which is what a complaint, a right to information request or a petition will need to cite.
  • A free place in a neighbourhood school until elementary education is complete (s3(1))
  • Admission without a capitation fee, an entrance test or a parent interview (s13(1))
  • A school where the child is not beaten or harassed (s17(1))
  • No expulsion before elementary education is complete (s16(4), as amended in 2019)
  • A seat in a private unaided school under the 25% entry quota, if the family falls in a notified group (s12(1)(c))
  • A say in a government or aided school through its School Management Committee (s21)
  • Teachers who attend regularly and complete the curriculum (s24(1))
  • Remedial teaching and a re-examination before any child is held back in class 5 or 8 (s16(2))
Article 51A(k) also places a duty on the parent to provide opportunities for education. Rights and duties sit side by side in the constitution, and officials sometimes cite the second to deflect the first.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series15www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Input norms
The RTE Schedule sets inputs, and says little about outcomes
The Schedule to the RTE Act lists what every school must have. Most entries are inputs that an inspector can count: teachers, classrooms, toilets, working days. None of them measures what a child has learned.
NormClasses I-VClasses VI-VIII
TeachersTwo for up to 60 children; ratio not above 40:1 beyond 200At least one per class; subject teachers for science and maths, social studies and languages
Working days a year200220
Instructional hours a year8001,000
Input norms were a reasonable first step in 2009, when many schools lacked classrooms, toilets and teachers. They are easy to verify and easy to fund.
The risk is that a school meeting every norm can still leave children unable to read. Sections 4 and 8 look at how India has tried to add learning to the checklist.
Source: Right of Children to Free and Compulsory Education Act 2009 (No. 35 of 2009), Schedule, India Code text.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series16www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
No-detention
Automatic promotion, its repeal and what replaced it
The original section 16 of the RTE Act said no child admitted to a school would be held back in any class or expelled until completing elementary education. Critics blamed it for weak effort; supporters said holding a child back without extra teaching only pushes her out of school.
RTE (Amendment) Act 2019
Act No. 1 of 2019, assent 10 January 2019, substituted section 16. There is now a regular examination in class 5 and class 8. A child who fails gets additional instruction and a re-examination within two months. The appropriate government may allow schools to hold her back, or may decide not to.
What stays
New section 16(4): "No child shall be expelled from a school till the completion of elementary education." In December 2024 the Centre amended its own RTE rules to use the power for schools under it (Careers360, 23 December 2024).
Detention is now a state choice. A practitioner should check the state's rules before assuming either regime applies.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series17www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
The courts
Six judgments that shape the right to education
The Supreme Court found a right to education before Parliament wrote one, and has since decided who the RTE Act binds and how hard the state must work to implement it. Each case below is still cited.
CaseDateWhat it decided
Mohini Jain v State of Karnataka30 July 1992Two judges: the right to education flows from the right to life in Article 21; capitation fees violate Article 14
Unni Krishnan J.P. v State of Andhra Pradesh4 February 1993Five judges: every child has a right to free education until 14; beyond that the right depends on the state's economic capacity
Society for Unaided Private Schools of Rajasthan v Union of India12 April 2012Three judges, 2-1, upheld s12(1)(c) as a reasonable restriction under Article 19(6); unaided minority schools exempted
Pramati Educational and Cultural Trust v Union of India2014Five judges exempted all minority schools, aided and unaided, from the RTE Act
Anjuman Ishaat-e-Taleem Trust v State of Maharashtra1 September 2025Two judges doubted Pramati and referred it to the Chief Justice for a larger bench
Dinesh Biwaji Ashtikar v State of Maharashtra13 January 2026Directed governments to frame rules under s38 for s12(1)(c) admissions and the NCPCR to file an affidavit on them; next listed 27 October 2026
Sources: the judgments and orders on Indian Kanoon. On 30 September 2026 a three-judge bench sent the Pramati question to the Chief Justice, because Pramati was decided by five judges. It was pending as of October 2026.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series18www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
03
Section Three
From Kothari to NEP 2020
CC BY-NC-ND 4.0 · ImpactMojo 101 Series19www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Kothari Commission
The Education Commission, 1964-66
The National Education Commission, chaired by D. S. Kothari, was formed on 14 July 1964 and submitted its report on 29 June 1966. It remains the most cited document in Indian education policy, and both later national policies borrow its targets.
  • A uniform 10+2+3 structure of schooling and higher education
  • A common school system of public education, vocationalised in general and special streams
  • Work experience and social or national service as part of education
  • Raising education spending from 2.9% of GDP to 6% by 1985-86
Why it still matters
The 6% figure reappears in 1968, 1986, 1992 and 2020. The common school idea is invoked by the Supreme Court in RTE cases, most recently in 2026.
Source: Wikipedia, Kothari Commission, summarising the 1966 report; the Supreme Court's judgment in Dinesh Biwaji Ashtikar (13 January 2026) cites the commission's common school system.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series20www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
NPE 1968
The first National Policy on Education
The 1968 National Policy on Education, adopted under Prime Minister Indira Gandhi, turned many of the Kothari recommendations into national policy. It was a statement of intent: education was then a state subject and there was no statutory right.
CommitmentWhat it meant in practice
Compulsory education to age 14A goal, later carried into Article 45 debates and finally into Article 21A in 2002
Three-language formulaHindi, English and the regional language; still contested in several states
Equal educational opportunityCalled for radical restructuring of the system
6% of national incomeKothari's spending target, adopted as policy
Source: Wikipedia, National Policy on Education. NEP 2020 (para 26.1) states that the 6% target was "envisaged by the 1968 Policy, reiterated in the Policy of 1986, and ... further reaffirmed in the 1992 review".
CC BY-NC-ND 4.0 · ImpactMojo 101 Series21www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
NPE 1986 and 1992
Equity, Operation Blackboard and the Programme of Action
The 1986 policy, under Rajiv Gandhi, put special emphasis on removing disparities for women, Scheduled Castes and Scheduled Tribes. It introduced Operation Blackboard to equip primary schools, a child-centred approach in primary education, and an expanded open university system through IGNOU. In 1992 the government under P. V. Narasimha Rao revised it through a Programme of Action.
What 1986 added
A focus on disadvantaged groups, minimum facilities in every primary school, and distance education for adults and higher education.
What it left open
Like 1968 it was a policy without a statutory right behind it. NEP 2020 calls its unfinished agenda "NPE 1986/92" and says it "is appropriately dealt with in this Policy".
Three decades separate the 1986 policy and NEP 2020. In between, most change came through schemes and new law.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series22www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Schemes
DPEP, Sarva Shiksha Abhiyan and Samagra Shiksha
Policy statements set direction; centrally sponsored schemes moved the money. The District Primary Education Programme (DPEP) was funded 85% by the Centre and 15% by states. Sarva Shiksha Abhiyan (SSA), launched in 2001, aimed to universalise elementary education, open schools in habitations without one and pull working children into class.
01
DPEP: district plans, 85:15 funding
→
02
SSA 2001: universal elementary education
→
03
RMSA: secondary schooling
→
04
Samagra Shiksha 2018: one integrated scheme
In 2018 SSA, Rashtriya Madhyamik Shiksha Abhiyan and teacher education schemes were merged into Samagra Shiksha, which now funds school education from pre-school to class 12.
PRS reports that the Centre and states share Samagra Shiksha funds 60:40 in most states and 90:10 in Himalayan and north-eastern states. Its 2026-27 allocation is Rs 42,100 crore.
Sources: Wikipedia, Sarva Shiksha Abhiyan; PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series23www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Timeline
Sixty years of Indian education policy on one page
The table puts the main policy, legal and measurement milestones in order. Notice the shift in the last decade from documents about access to instruments that measure learning.
YearMilestone
1964-66Kothari Commission; report submitted 29 June 1966
1968First National Policy on Education
197642nd Amendment moves education to the Concurrent List (in force 3 January 1977)
1986 / 1992Second NPE; Programme of Action revision
2001Sarva Shiksha Abhiyan launched
200286th Amendment inserts Article 21A
2009 / 2010RTE Act passed; Article 21A and the Act in force 1 April 2010
2018Samagra Shiksha merges SSA and RMSA
2019RTE amendment replaces automatic promotion
2020NEP 2020 approved by the Union Cabinet on 29 July 2020
2021NIPUN Bharat foundational learning mission launched
2024PARAKH Rashtriya Sarvekshan, the renamed National Achievement Survey
2025VBSA Bill to replace UGC, AICTE and NCTE introduced, 15 December
CC BY-NC-ND 4.0 · ImpactMojo 101 Series24www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
NEP 2020 structure
NEP 2020 replaces 10+2 with 5+3+3+4
NEP 2020 states that "the extant 10+2 structure in school education will be modified with a new pedagogical and curricular restructuring of 5+3+3+4 covering ages 3-18". The change brings three years of pre-school or anganwadi into the school system for the first time.
StageYearsGradesAges
Foundational53 years pre-school or anganwadi + Grades 1-23-8
Preparatory3Grades 3-58-11
Middle3Grades 6-811-14
Secondary4Grades 9-12, in two phases14-18
The foundational stage is meant to be "flexible, multilevel, play/activity-based". Light textbooks start only in the preparatory stage.
The medium of instruction until at least Grade 5, but preferably till Grade 8 and beyond, should be the home language or mother tongue "wherever possible".
Source: National Education Policy 2020, Ministry of Education, paras 4.1-4.2 and 4.11.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series25www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
NEP 2020 targets
The numbers NEP 2020 committed to
NEP 2020 is unusually specific about targets. Some have dates that have now passed, which makes them useful benchmarks for judging delivery.
5 crore+
elementary students estimated not to have foundational literacy and numeracy (NEP para 2.1)
National Education Policy 2020, Ministry of Education
3.22 crore
out-of-school children aged 6-17, NSS 75th round, 2017-18 (NEP para 3.1)
National Education Policy 2020, Ministry of Education
100% GER
target from pre-school to secondary by 2030
National Education Policy 2020, Ministry of Education
50% by 2035
target higher education GER, up from 26.3% in 2018
National Education Policy 2020, Ministry of Education
Universal foundational literacy and numeracy in primary school was the "highest priority", to be achieved by 2025. The NIPUN Bharat mission later set the target as 2026-27 (PRS, February 2026).
Public spending on education was to reach 6% of GDP "at the earliest", from "around 4.43%" in 2017-18 budget estimates (NEP para 26.1).
CC BY-NC-ND 4.0 · ImpactMojo 101 Series26www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Policy versus law
NEP 2020 is a policy, so it binds only through other instruments
NEP 2020 was approved by the Union Cabinet. It is not an Act of Parliament and does not amend the RTE Act. Each element has to travel through some other instrument before it binds anyone: a statute, a set of rules, a scheme guideline, a board regulation or a state government order.
Where NEP has legal teeth
Where the RTE Act or rules were amended (for example the class 5 and 8 examinations), where a scheme makes funds conditional, or where a regulator such as a school board changes its own rules.
Where it does not
The age-3 start, mother-tongue instruction and the three-language formula depend on each state. States can and do adopt parts of NEP and decline others, because education is on the Concurrent List.
When someone says "NEP requires", ask which document turns that requirement into a rule, and whether the state you work in has adopted it.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series27www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
04
Section Four
Access versus learning: the data
CC BY-NC-ND 4.0 · ImpactMojo 101 Series28www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Enrolment
Almost every child aged 6 to 14 is enrolled
ASER is a household survey run by Pratham and local partners. ASER 2024 reached 649,491 children in 17,997 villages across 605 rural districts. It finds that enrolment for children aged 6 to 14 has been above 95% for close to 20 years. The access problem that dominated the 1990s is now largely solved for this age group.
98.1%
of rural children aged 6-14 enrolled in school, 2024
ASER Centre, ASER 2024 National findings (rural India), released January 2025
66.8%
of rural 6-14 year olds in government schools, 2024 (72.9% in 2022)
ASER Centre, ASER 2024 National findings (rural India), released January 2025
77.4%
of 3-year-olds in some pre-primary institution, up from 68.1% in 2018
ASER Centre, ASER 2024 National findings (rural India), released January 2025
7.9%
of 15-16 year olds not enrolled, 2024 (13.1% in 2018)
ASER Centre, ASER 2024 National findings (rural India), released January 2025
The 2022 jump in government enrolment came during the pandemic and has partly reversed. Watch the 15-16 age group: girls not enrolled exceed 10% in Madhya Pradesh (16.1%), Uttar Pradesh (15%) and Rajasthan (12.7%).
CC BY-NC-ND 4.0 · ImpactMojo 101 Series29www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
UDISE+
The school system in numbers, 2025-26
UDISE+ is the Ministry of Education's census of schools. Since 2022-23 it collects individual student records uploaded by schools, so figures from 2022-23 onward are not comparable with earlier years. The 2025-26 report was released on 7 July 2026.
Indicator2024-252025-26
Total schools14,71,47314,66,682
Government schools10,13,32210,05,245
Private unaided schools3,39,5833,41,689
Enrolment, Grades I-XII23,28,85,60023,23,06,857
Teachers1,01,22,4201,02,73,020
Single-teacher schools1,04,1251,00,843
Zero-enrolment schools7,9935,663
Source: UDISE+ 2025-26, Ministry of Education, released 7 July 2026 and UDISE+ 2024-25 reports, Ministry of Education, Table 1 and Table 2.2. Including pre-primary, 2025-26 enrolment was 24.72 crore.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series30www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Enrolment ratios
Primary enrolment is falling and secondary is rising
Gross enrolment ratio (GER) divides everyone enrolled at a level by the population of the official age for that level, so it can exceed 100. Net enrolment ratio (NER) counts only children of the right age.
LevelGER 24-25GER 25-26NER 25-26
Primary90.989.477.1
Upper primary90.389.668.4
Secondary78.781.550.8
Higher secondary58.461.739.0
Primary enrolment (Grades I-V) fell 2.36% in one year, to 10.19 crore. Secondary and higher secondary rose 3.0% and 4.4%.
A.C. Mehta (July 2026) reads the primary fall as consistent with falling births and a smaller primary-age population, which means fewer children per school and more small schools.
Source: UDISE+ 2025-26, Ministry of Education, released 7 July 2026 and UDISE+ 2024-25 reports, Ministry of Education (NER from Table 6.4). The higher secondary GER of 61.7 is far below NEP's 100% target for 2030.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series31www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Dropout
Who leaves school, and why
1.8%
preparatory stage dropout rate, 2025-26
UDISE+ 2025-26, Ministry of Education, released 7 July 2026
3.6%
middle stage dropout rate, 2025-26 (3.5% in 2024-25)
UDISE+ 2025-26, Ministry of Education, released 7 July 2026
7.0%
secondary stage dropout rate, 2025-26
UDISE+ 2025-26, Ministry of Education, released 7 July 2026
51.9%
secondary retention rate, up from 47.2%
UDISE+ 2025-26, Ministry of Education, released 7 July 2026
Reason for not attending (age 14-18)Share
Supplement household income44%
Domestic chores28%
Education not necessary8%
School too far1%
Other19%
Source: unit-level PLFS 2023-24, Economic Survey 2025-26, as reported by PRS (February 2026).
Distance is rarely the reason given now. Work and household duties are, which points to cash transfers, flexible schooling and skilling at secondary level more than new buildings. PRS notes the Economic Survey 2025-26 recommended integrating skilling into secondary schools.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series32www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
ASER reading
Reading has recovered past its pre-pandemic level
ASER tests every sampled child aged 5 to 16 one-on-one at home, using the same method since 2006. The key indicator is whether a child can read a story at Standard II level of difficulty.
Government school children who can read a Std II level text (%), rural India
ASER Centre, ASER 2024 National findings (rural India), released January 2025
The recovery is real, and so is the level: in 2024 more than three in four government school children in Std III still could not read a Std II text. In private schools 59.3% of Std V children could, against 44.8% in government schools.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series33www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
ASER arithmetic
Arithmetic is at its highest in a decade, from a low base
ASER's arithmetic tasks check whether a child can recognise numbers, do a two-digit subtraction with borrowing, and divide a three-digit number by a one-digit number. ASER 2024 reports that basic arithmetic reached its highest level in over a decade in both government and private schools.
33.7%
of Std III children can at least do subtraction, all schools (28.2% in 2018)
ASER Centre, ASER 2024 National findings (rural India), released January 2025
27.6%
the same for Std III in government schools (20.2% in 2022)
ASER Centre, ASER 2024 National findings (rural India), released January 2025
30.7%
of Std V children can do division (27.9% in 2018)
ASER Centre, ASER 2024 National findings (rural India), released January 2025
45.8%
of Std VIII children can do division (44.1% in 2018)
ASER Centre, ASER 2024 National findings (rural India), released January 2025
Gains are concentrated in the early grades, where the foundational learning mission has focused since 2021.
Std VIII barely moved. More than half of children about to finish elementary school cannot divide a three-digit number by a one-digit number.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series34www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
PARAKH 2024
The government's own survey shows scores falling with each stage
PARAKH Rashtriya Sarvekshan 2024, the successor to the National Achievement Survey, tested 21,15,022 students in Grades 3, 6 and 9 on 4 December 2024, in over 74,000 government, aided, private and central schools in 781 districts.
Average score (out of 100) by grade, PARAKH Rashtriya Sarvekshan 2024
PARAKH Rashtriya Sarvekshan 2024, National Report (NCERT, 2025)
The same report gives Grade 9 state government school averages of 48 in language and 33 in mathematics, against 69 and 48 in central government schools.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series35www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Reading two surveys
ASER and PARAKH measure different things on purpose
Practitioners often see ASER and PARAKH numbers quoted side by side as if they measure the same thing. They do not use the same sample, method or bar, so a difference between them is not by itself a sign that one is wrong.
FeatureASER 2024PARAKH Rashtriya Sarvekshan 2024
Run byPratham and local partnersPARAKH at NCERT, Ministry of Education
WhereHouseholds, rural districts onlySchools, all 36 states and UTs
WhoChildren aged 3-16, in or out of schoolStudents in Grades 3, 6 and 9
TestOne-on-one, basic reading and arithmeticWritten, competency-based test booklets
ReportsShare of children at each levelAverage percentage score
Scale649,491 children21,15,022 students
Use ASER for foundational skills and out-of-school children; use PARAKH for curriculum-linked competencies and comparisons across school types and states.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series36www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
05
Section Five
Paying for school
CC BY-NC-ND 4.0 · ImpactMojo 101 Series37www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
The 6% target
Sixty years of a spending target that has not been met
Every Indian education policy since 1966 has named the same target: public spending on education of 6% of GDP. Kothari recorded a starting point of 2.9%. NEP 2020 put the level at around 4.43% in 2017-18. The PRS analysis of the 2026-27 budget estimates combined Centre and state spending at 4.1% of GDP in 2022-23.
2.9%
of GDP when the Kothari Commission reported, 1966
Wikipedia, Kothari Commission
4.43%
Centre plus states, 2017-18 budget estimates
National Education Policy 2020, Ministry of Education
4.1%
Centre plus states, 2022-23
PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026
6%
target in 1966, 1968, 1986, 1992 and 2020
National Education Policy 2020, Ministry of Education
PRS compares India's 4.1% (2022) with Germany 5.2% (2022), and the US 5.4% and the UK 5.9% (2021), from World Bank data. China is 4.0% (2023).
A percentage of GDP is a blunt measure. A rising GDP and a falling child population both change what a given share buys per pupil.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series38www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Union budget 2026-27
The Ministry of Education's budget for 2026-27
The Union Budget 2026-27, presented on 1 February 2026, allocated Rs 1,39,289 crore to the Ministry of Education, 14% more than the 2025-26 revised estimate. School education took 60% and higher education 40%.
Head2024-25 actual2025-26 RE2026-27 BE
School education65,15970,56783,562
of which Samagra Shiksha36,50238,00042,100
of which PM POSHAN (midday meals)9,90310,60012,750
of which PM SHRI schools3,5044,5007,500
Higher education45,57751,38255,727
of which central universities16,04217,08517,440
of which IITs10,30911,52512,123
Total, Ministry of Education1,10,7361,21,9491,39,289
Rs crore. Source: PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026, from Demands No. 25 and 26, Expenditure Budget 2026-27.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series39www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Where the money goes
Samagra Shiksha and higher education dominate
The Union budget is only part of public education spending, because states pay most teacher salaries. Within the Union's share, three lines stand out: Samagra Shiksha, the midday meal and the central institutions of higher education.
Ministry of Education allocation 2026-27 (Rs crore)
PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026
PRS estimates education at around 2.4% to 2.6% of the overall Union budget between 2024-25 and 2026-27. Across all levels of government, WDI puts education at 14.2% of government expenditure in India in 2022.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series40www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Fiscal federalism
Who pays for a government school
A government school in Bihar or Tamil Nadu is funded by several streams at once. Most of the cost is teacher salaries paid from the state budget. The Centre adds money through centrally sponsored schemes that the state must match.
01
State budget: salaries, most recurrent cost
→
02
Samagra Shiksha: 60:40 (90:10 in Himalayan and north-eastern states)
→
03
PM POSHAN: midday meal, pre-primary to Grade 8
→
04
PM SHRI: about 14,500 model schools, 2022-23 to 2027-28
PM POSHAN, launched in 2021-22, subsumed the midday meal scheme and covers nearly 11.2 crore children in government and aided schools.
PM SHRI's central share for 2022-23 to 2027-28 is Rs 18,128 crore and the state share Rs 9,232 crore; 13,070 schools had been upgraded by January 2026.
Source: PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026. Matching grants mean a state that cannot find its 40% leaves central money unspent.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series41www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Spending the money
Allocations are only half the story
A budget line shows intent. Whether the money reaches schools depends on release, spending and timing. The PRS analysis records several gaps that matter to anyone monitoring a scheme.
FindingDetail
Late spendingBy February 2025 the school education department had spent about 59% of its allocation, leaving the rest for the last two months
Committee adviceStanding Committee (2025): last-quarter spending at most 33% of the allocation, and at most 15% in the final month
NIPUN BharatRs 7,178 crore approved for 2021-22 to 2023-24; Rs 5,007 crore (70%) spent
PM USHA (higher education)On average 16% of allocated funds spent between 2022-23 and 2024-25
Teacher trainingMinistry spending fell from Rs 599 crore (2019-20) to Rs 96 crore (2022-23)
Source: PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026. A March rush of spending usually buys goods and events. It rarely buys the year-round teaching support a learning programme needs.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series42www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Household spending
Families pay even in free government schools
The National Statistics Office ran a Comprehensive Modular Survey on Education from April to June 2025. PRS summarises what households told it about the cost of school, which matters because the RTE Act promises free education to age 14.
10x
cost of a private school relative to a government school, 2025
PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026, citing NSO CMS: Education 2025
27%
of government school students report paying course fees at some level
PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026, citing NSO CMS: Education 2025
Rs 2,863
average annual cost per student in a government school (fees and transport)
PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026, citing NSO CMS: Education 2025
9.6 crore
children in private schools, 2024-25
PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026, citing UDISE+ 2024-25
About 8% of government school students in Grades 1 to 8 pay an average course fee of Rs 229 a year, inside the RTE age band.
Private enrolment kept rising despite the tenfold cost gap, which tells you parents are paying for something they value. Section 7 asks what.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series43www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Money and learning
More inputs alone rarely raise learning
The 2023 report of the Global Education Evidence Advisory Panel (GEEAP) lists "funding additional inputs alone, when other issues are not addressed" as a bad buy. Its examples are textbooks, additional teachers to reduce class size, buildings, grants, salary and libraries.
The fiscal cost of absence
Muralidharan, Das, Holla and Mohpal (2017) found 23.6% of teachers absent in unannounced visits to 1,297 villages and put the salary cost of unauthorised absence at $1.5 billion a year.
Where money works
The same study estimates better monitoring could be over ten times more cost effective at raising the effective pupil-teacher ratio than hiring more teachers.
The argument is about the next rupee. Inputs matter where they are missing; where they exist, how they are used decides learning. Sources: Global Education Evidence Advisory Panel, 2023 Cost-Effective Approaches to Improve Global Learning; Journal of Public Economics 145 (2017).
CC BY-NC-ND 4.0 · ImpactMojo 101 Series44www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
06
Section Six
Teachers
CC BY-NC-ND 4.0 · ImpactMojo 101 Series45www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
The workforce
India now has more than one crore school teachers
UDISE+ 2024-25 recorded, for the first time, more than one crore teachers. By 2025-26 the count was 1,02,73,020. Teachers are the largest item in every state education budget, so how they are recruited, posted and supported is the biggest spending decision in education policy.
1.03 crore
school teachers, 2025-26
UDISE+ 2025-26, Ministry of Education, released 7 July 2026
54.9%
of teachers are women, 2025-26
UDISE+ 2025-26, Ministry of Education, released 7 July 2026
10 : 1
pupil-teacher ratio, foundational stage
UDISE+ 2025-26, Ministry of Education, released 7 July 2026
21 : 1
pupil-teacher ratio, secondary stage
UDISE+ 2025-26, Ministry of Education, released 7 July 2026
National averages look comfortable against the RTE Schedule's ceiling of 40 pupils per teacher at primary level. The problem lies in distribution.
PRS reports higher secondary ratios of 47:1 in Jharkhand, 37:1 in Maharashtra and Odisha, and 35:1 in Uttar Pradesh in 2024-25.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series46www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Vacancies
Ten lakh posts vacant
NEP 2020 says teacher vacancies "will be filled at the earliest, in a time-bound manner", especially in disadvantaged areas. Six years later, PRS reports that nearly 10 lakh teaching posts were vacant in 2024-25, citing the Rajya Sabha Standing Committee on Education's 368th report (August 2025).
  • Lack of regular recruitment
  • Posts not sanctioned
  • Shortage of subject-specialist teachers
  • Small schools that make distribution hard
Contract teachers
The Standing Committee noted that contractual recruitment in central government schools such as Kendriya Vidyalayas nearly doubled between 2023-24 and 2024-25, and recommended filling vacancies through regular appointment.
Reasons as listed by PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026. Vacancy counts depend on sanctioned posts, which states set themselves, so compare states with care.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series47www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Small schools
One lakh schools still have a single teacher
A single-teacher school asks one adult to teach several grades at once, keep records, run the midday meal and attend training. When that teacher is absent the school is closed. UDISE+ shows the number falling steadily, partly through consolidation.
YearSingle-teacher schoolsZero-enrolment schools
2022-231,18,19010,294
2023-241,10,97112,954
2024-251,04,1257,993
2025-261,00,8435,663
Falling primary enrolment means more schools with few children. Merging them can free teachers but lengthens journeys for the youngest children.
Government schools fell by 8,077 in 2025-26 while private unaided schools rose by 2,106.
Sources: PIB release on UDISE+ 2024-25, 28 August 2025 (2022-23 and 2023-24); UDISE+ 2025-26, Ministry of Education, released 7 July 2026 and UDISE+ 2024-25 reports, Ministry of Education.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series48www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Absence in 2003
One in four teachers absent on any given day
Kremer, Chaudhury, Rogers, Muralidharan and Hammer (2005) made surprise visits to government primary schools across India. Their snapshot, published in the Journal of the European Economic Association, became the reference figure for teacher absence in South Asia.
25%
of teachers absent from school during unannounced visits
Kremer et al. (2005), JEEA 3(2-3)
15%
absence in Maharashtra, the lowest state
Kremer et al. (2005)
42%
absence in Jharkhand, the highest state
Kremer et al. (2005)
Lower absence where
Schools had been inspected recently, had better infrastructure and were closer to a paved road.
No difference where
Teachers were paid more, were locally recruited, or the school had a parent association.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series49www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Absence a decade on
Absence fell only slightly, and the cost is large
Muralidharan, Das, Holla and Mohpal (2017) revisited a nationally representative panel of 1,297 villages. Absence had fallen only a little, to 23.6%, while pupil-teacher ratios had improved. Their paper links the two: hiring more teachers and reducing pupil-teacher ratios was associated with higher absence.
23.6%
teachers absent in unannounced visits
Muralidharan et al. (2017), JPubE 145
$1.5 bn
annual salary cost of unauthorised absence
Muralidharan et al. (2017)
More frequent monitoring was strongly associated with lower absence.
The authors estimate that better monitoring could be over ten times more cost effective at raising the effective pupil-teacher ratio than hiring additional teachers.
This is correlational evidence from a panel. The next slide turns to experiments.
Absence is a management failure more than a moral one: it falls where someone checks.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series50www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Incentives
Two experiments on teacher incentives
Two randomised experiments in India tested whether changing teachers' incentives changes what children learn. Both found that it does, at modest cost.
StudySetting and designResult
Duflo, Hanna and Ryan (2012), AER 102(4)60 informal one-teacher schools in rural India; a child photographed the teacher and students at the start and end of each day with a tamper-proof camera; pay tied to attendanceAbsence fell from 42% in comparison schools to 22%; test scores 0.17 standard deviations higher after one year
Muralidharan and Sundararaman (2011), JPE 119(1)Government rural primary schools in Andhra Pradesh; bonus tied to students' test score gains, mean bonus 3% of annual payAfter two years, 0.28 SD in maths and 0.16 SD in language; gains also in subjects without incentives
Effects are in standard deviations (SD) of the test score distribution, so they can be compared across tests. The performance pay schools also beat schools given extra inputs of similar value.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series51www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Qualifications and training
Many teachers lack professional qualifications
Section 23 of the RTE Act requires teachers to hold the minimum qualifications set by an academic authority; the National Council for Teacher Education sets them. UDISE+ 2024-25, as reported by PRS, shows the gap by level.
LevelProfessionally unqualified
Pre-primary52%
Grades 1-512%
Grades 6-812%
Grades 9-1010%
Grades 11-1211%
Qualification is not the same as competence, but an untrained teacher of four-year-olds is a direct risk to NEP's foundational stage.
NISHTHA, the in-service training programme launched in 2019 under Samagra Shiksha, had trained 43% of targeted teachers and 49% of targeted school heads by January 2026 (PRS).
In Anjuman Ishaat-e-Taleem Trust (1 September 2025) the Supreme Court held the TET binding on teachers recruited before the RTE Act: those with more than five years to retire must pass it within two years or leave service; those with less may stay without it but cannot be promoted (paras 216-217).
Source: PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026. Ministry spending on teacher training fell from Rs 599 crore in 2019-20 to Rs 96 crore in 2022-23.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series52www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Community teachers
Local tutors can teach the basics well
Banerjee, Cole, Duflo and Linden (2007) evaluated Pratham's Balsakhi programme, which hired young women from the community to teach basic literacy and numeracy to children who had fallen behind. These tutors had far less training than regular teachers.
0.28 SD
rise in average test scores of all children in treatment schools
Banerjee et al. (2007), QJE 122(3)
0.10 SD
gain remaining for targeted children a year after the programme ended
Banerjee et al. (2007)
The gains were largest for the weakest students, which is the group a flat learning profile leaves behind.
GEEAP (2023) rates "augmenting teaching teams with community-hired staff" as promising but with limited evidence on cost-effectiveness at scale.
For policy the lesson is targeted: a tutor teaching at the child's level can work, but it is a complement to regular teachers.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series53www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
07
Section Seven
Private schools and section 12(1)(c)
CC BY-NC-ND 4.0 · ImpactMojo 101 Series54www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
The private shift
Private unaided schools now teach 38% of students
UDISE+ 2025-26 shows government schools' share of Grades I-XII enrolment slipping below half, while private unaided schools gained about 25 lakh students in a year in which total enrolment fell.
Share of Grades I-XII enrolment by school management, 2025-26 (%)
A.C. Mehta, analysis of UDISE+ 2025-26 (July 2026), from UDISE+ 2025-26
In 2024-25 the government share was 50.83% and private unaided 36.80%. Rural figures differ: ASER 2024 finds 66.8% of rural children aged 6-14 in government schools.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series55www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Why parents choose
What parents say they pay for
The NSSO survey of household social consumption on education (2017-18) asked students in private institutions why they chose them. PRS reproduces the answers. Quality and proximity dominate, with English medium a strong third.
Reason givenShare of respondents
Quality of public institution not satisfactory34%
Private institution located nearby27%
Uses English as medium of instruction17%
Facilities such as transport and hostels14%
Perceived quality and English are both signals that a parent can see. Learning levels are much harder for a parent to observe.
That information gap is why report cards and published assessment results (Section 8) can change school markets.
Source: PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026, citing NSSO, Household Social Consumption on Education in India, 2017-18.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series56www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Private versus public
Raw gaps overstate what private schools add
ASER 2024 finds 59.3% of Std V children in rural private schools can read a Std II text, against 44.8% in government schools. That gap mixes two things: what the school adds, and which families choose private schools in the first place.
Selection
Families who pay fees tend to be better off and more educated, with more books and help at home. Their children would score higher in any school.
Value added
To isolate the school's contribution you need a design that compares similar children, such as a lottery. The next slide describes one from Andhra Pradesh.
Selection bias
A difference between groups caused by who chooses to join each group, mistaken for the effect of the group itself. See Causal Inference 101.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series57www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
A school voucher experiment
Evidence from a school voucher lottery in Andhra Pradesh
Muralidharan and Sundararaman (2015, QJE 130(3)) studied a programme in Andhra Pradesh that gave students vouchers to attend a private school of their choice. A two-stage lottery, first across villages and then across applicants, let them measure both the effect on winners and spillovers on everyone else.
No gap
in Telugu and maths test scores between lottery winners and losers after two and four years
Muralidharan and Sundararaman (2015)
Under 1/3
the cost per student in private schools, relative to public schools in the sample
Muralidharan and Sundararaman (2015)
No spillover
harm to public school students or to private school students
Muralidharan and Sundararaman (2015)
The authors conclude private schools in this setting delivered slightly better test score gains than public ones at much lower cost per student. Test scores are one outcome; the study does not settle questions of equity or segregation.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series58www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
How 12(1)(c) works
A quarter of entry seats in private schools for disadvantaged children
Section 12(1)(c) of the RTE Act requires unaided schools to admit, in class I (or the entry class), "to the extent of at least twenty-five per cent. of the strength of that class", children from weaker sections and disadvantaged groups in the neighbourhood, and to provide free education till elementary completion.
01
STATE notifies weaker sections and disadvantaged groups
→
02
SCHOOL declares entry seats
→
03
LOTTERY or portal allots seats
→
04
STATE reimburses the school under s12(2)
Under s12(2) the state reimburses the lower of its own per-child expenditure and the school's actual fee.
Section 13(1) bars capitation fees and screening, so a school cannot test a child or interview parents to fill these seats.
Source: Right of Children to Free and Compulsory Education Act 2009 (No. 35 of 2009), ss 12-13. In 2012 the Supreme Court upheld the provision as a reasonable restriction on the freedom to run a business.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series59www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Implementation gaps
Sixteen years on, the Supreme Court is still asking for compliance
In Dinesh Biwaji Ashtikar v State of Maharashtra (13 January 2026) the Supreme Court held that without enforceable rules "the object of Article 21A and the statutory policy under Section 12(1)(c) would be a dead letter". It identified practical barriers that a practitioner will recognise.
  • Online application portals that assume digital literacy
  • Forms and notices in languages parents do not read
  • No public list of how many seats each school has
  • Refusals without written reasons
  • Rules under s38 framed with NCPCR and state commissions (a direction)
  • NCPCR to collate the rules and file an affidavit by 31 March 2026
  • Seats published before applications open (guideline)
  • Denials recorded with reasons and reviewed by the Block Education Officer within 72 hours (guideline)
  • Help-desks and a window to correct defective forms (guideline)
Source: the judgment, paras 13-16. On 29 September 2026 the Court asked the NCPCR chairperson to appear and listed the case on 27 October 2026; check the latest order before relying on it.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series60www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Regulating private schools
Four levers a state can pull
Once private schools educate four in ten children, regulating them well becomes education policy. States have used four broad levers, and each has a cost.
LeverWhat it doesRisk
Recognition normsMinimum land, buildings and teacher norms before a school can operateLow-fee schools close or operate unrecognised
Fee regulationCaps or approvals for fee increasesSchools cut spending or add hidden charges
Seat mandatess12(1)(c) admissions with reimbursementLate or low reimbursement turns schools against the scheme
InformationPublishing learning results for every schoolTeaching to the test if the measure is narrow
Information has the clearest evidence. Andrabi, Das and Khwaja (2017, AER 107(6)) gave report cards to parents in Pakistani villages: test scores rose 0.11 SD, private fees fell 17% and primary enrolment rose 4.5%.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series61www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
08
Section Eight
Accountability and assessment
CC BY-NC-ND 4.0 · ImpactMojo 101 Series62www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Accountability
Three ingredients of accountability in a school system
WDR 2018 asks systems to "assess learning" and "act on evidence". Making a school system answer for learning needs three things that are often missing at the same time: someone has to know how children are doing, someone has to care, and someone has to be able to change what happens.
01
INFORMATION: credible data on learning
→
02
INCENTIVES: consequences or recognition tied to it
→
03
CAPACITY: skills and time to respond
→
04
FEEDBACK: the next measurement shows whether it worked
Information without capacity
A head teacher told her school is weak, with no support or authority to change anything, can only file the report.
Incentives without information
Rewards tied to enrolment or attendance registers push schools to improve the register, which is easier than improving the learning.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series63www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
National assessment
From NAS to PARAKH Rashtriya Sarvekshan
NEP 2020 (para 4.41) proposed a national assessment centre, PARAKH, as a standard-setting body that would also run the National Achievement Survey and guide State Achievement Surveys. PARAKH sits within NCERT. Its 2024 survey is described as the large-scale assessment of India, also known as NAS.
GradeLanguage 2017Language 2021Maths 2017Maths 2021
367626357
558555344
856534236
1036433432
Average scores out of 100, National Achievement Survey 2017 and 2021, as reported by PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026. Scores fell between 2017 and 2021 in grades 3, 5 and 8 in both subjects, spanning the pandemic school closures.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series64www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
School management committees
Parents on the committee do not guarantee a voice
Section 21 of the RTE Act requires every government and aided school to form a School Management Committee with local authority representatives, parents or guardians and teachers. The idea is the short route of accountability: parents watch the school directly instead of waiting for the state to act.
What the evidence says
Kremer et al. (2005) found that schools with parent associations had no lower teacher absence. GEEAP (2023) lists "involving communities in school management" as promising but with limited evidence.
What helps a committee work
Information parents can use (a simple reading test of their own children), authority over some money, and a clear channel to escalate complaints.
A committee that exists on paper satisfies s21. Whether it changes anything depends on what it knows and what it controls.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series65www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Information as a tool
Telling parents and teachers what children know
GEEAP (2023) rates "providing information on the benefits, costs, and quality of education" as a great buy, its highest category. The logic is simple: families decide how long a child stays in school and which school to pick, so better information changes decisions at very low cost.
+0.11 SD
test scores in villages that received school report cards
Andrabi, Das and Khwaja (2017), AER 107(6)
-17%
private school fees in those villages
Andrabi, Das and Khwaja (2017)
The Pakistan experiment randomly assigned half the sample villages to receive report cards showing schools' test results.
Weaker private schools improved or cut fees, and enrolment rose 4.5%.
ASER's citizen-led testing in India rests on the same idea at national scale: a number that anyone can understand.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series66www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
When reform becomes paperwork
A management reform that changed nothing
Muralidharan and Singh (NBER Working Paper 28129) evaluated a large school management programme in India: comprehensive school assessments, detailed school ratings and customised improvement plans in 1,774 schools. It did not change incentives or accountability.
1,774
schools in the experiment
Muralidharan and Singh, NBER w28129
No impact
on school functioning or student outcomes
Muralidharan and Singh, NBER w28129
600,000+
schools in the national scale-up; in the state studied it still did not raise learning
Muralidharan and Singh, NBER w28129
Interviews with officials found the main effect on the ground was more reporting and paperwork. Assessments were completed and the ratings were informative, so the programme looked successful on compliance measures.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series67www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Board exams
Lowering the stakes of board examinations
NEP 2020 (para 4.36) says board and entrance examinations force students to concentrate on a few subjects and reward coaching. It keeps Grade 10 and 12 boards but asks for them to test core capacities and to lower the stakes.
What NEP proposes
Students may take board exams on up to two occasions in a school year, one main and one for improvement. Boards may move to annual, semester or modular exams. School examinations in Grades 3, 5 and 8 track progress.
Trade-offs
Two attempts reduce the cost of one bad day but add administrative load. Exams in Grades 5 and 8 can now lead to detention (Section 2), so the same test carries higher stakes for younger children.
Source: National Education Policy 2020, Ministry of Education, paras 4.36-4.40. Implementation varies by board; check the board's current scheme before advising students.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series68www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Data quality
Know the limits of the numbers you cite
Education policy in India runs on administrative data. UDISE+ figures are uploaded by schools holding active codes, and the Ministry disclaims responsibility for errors in that self-reported data. Since 2022-23 the data are individual student records, so they cannot be compared with 2021-22 and earlier.
  • Registers count enrolment; attendance can be far lower
  • Dropout rates are derived from year-to-year records
  • GER depends on population projections, which age as the Census ages
  • Learning surveys differ in sample and method
Practical rules
Cite the year and source of every number. Do not join series across a method break. Prefer household surveys for out-of-school children. Triangulate a surprising figure against a second source.
The Census with reference date 1 March 2027 will reset the population denominators behind GER and NER. Source for UDISE+ caveats: UDISE+ 2025-26, Ministry of Education, released 7 July 2026 and UDISE+ 2024-25 reports, Ministry of Education.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series69www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
09
Section Nine
What works: the evidence
CC BY-NC-ND 4.0 · ImpactMojo 101 Series70www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Smart buys
The GEEAP ranking of education interventions
The Global Education Evidence Advisory Panel, convened by the UK FCDO, UNICEF, USAID and the World Bank, reviewed more than 550 evaluations for its 2023 report and sorted interventions by cost-effectiveness and evidence of working at scale. Its members include Rukmini Banerji of Pratham, Karthik Muralidharan and Tahir Andrabi.
CategoryExamples
Great buysInformation on benefits, costs and quality of education; structured pedagogy; teaching by learning level instead of grade
Good buysParent-directed early stimulation (0-36 months); quality pre-primary (3-5); reducing travel time; merit scholarships for disadvantaged children; school-based deworming where worm load is high
Promising, limited evidenceAdaptive learning software where hardware exists; community-hired staff; mobile phones; safeguarding; socio-emotional skills; community school management; targeting girls
Effective but relatively expensiveCash transfers (as a tool for learning); feeding in primary schools
Bad buysHardware alone; additional inputs alone, such as textbooks, more teachers, buildings, grants, salary and libraries
CC BY-NC-ND 4.0 · ImpactMojo 101 Series71www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Teaching at the Right Level
Grouping children by what they can do
Teaching at the Right Level (TaRL)
Pratham's approach: assess each child with a simple reading and arithmetic test, group children by level for part of the day regardless of grade, teach with level-appropriate activities, and reassess regularly.
01
ASSESS: one-on-one basic test
→
02
GROUP: by level, for part of the day
→
03
TEACH: activities for that level
→
04
REASSESS: move children up as they learn
Banerjee and colleagues (2017, Journal of Economic Perspectives 31(4)) describe a series of randomised trials in Indian schools that took the approach from a proof of concept to government scale. Some designs failed inside the regular school system, and those failures shaped the later versions; two versions eventually raised learning in government schools at scale.
GEEAP lists targeting instruction by learning level as a great buy, in or out of school.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series72www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Remedial and computer-assisted learning
Balsakhi: a tutor and a computer
Banerjee, Cole, Duflo and Linden (2007, QJE 122(3)) ran two randomised experiments in urban Indian schools. One hired community tutors (balsakhis) to teach children who were behind; the other gave children time on mathematics software.
0.28 SD
remedial tutoring: rise in average test scores in treatment schools
Banerjee, Cole, Duflo and Linden (2007)
0.47 SD
computer-assisted mathematics: rise in maths scores
Banerjee, Cole, Duflo and Linden (2007)
0.10 SD
what remained for targeted children a year after the programmes ended
Banerjee, Cole, Duflo and Linden (2007)
Both worked while running, and both benefited the weakest children most.
Fade-out is common. Programmes need to continue, or to change the regular classroom, if the gains are to last.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series73www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Mindspark
Adaptive software in urban India
Muralidharan, Singh and Ganimian (2019, AER 109(4)) studied Mindspark, a personalised, technology-aided after-school programme in urban India. Middle-school students were offered free access by lottery. The software adapts each question to the level the child has actually reached, which may be several grades below the one she is enrolled in.
0.37 SD
maths gain for lottery winners in 4.5 months
Muralidharan, Singh and Ganimian (2019)
0.23 SD
Hindi gain in the same period
Muralidharan, Singh and Ganimian (2019)
With full attendance of 90 days the authors project gains of 0.6 SD in maths and 0.39 SD in Hindi.
Absolute gains were similar for all students, so relative gains were much larger for academically weaker students.
GEEAP rates adaptive software as promising where hardware already exists, and hardware alone as a bad buy.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series74www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Structured pedagogy
Lesson plans, materials and coaching together
GEEAP defines structured pedagogy as "a package that includes structured lesson plans, learning materials, and ongoing teacher support" and rates it a great buy. Each element alone does less: a textbook without a plan, or training without follow-up, rarely changes daily teaching.
NIPUN Bharat
India's National Mission on Foundational Literacy and Numeracy, launched in 2021, aimed for every child to read and do basic arithmetic by Grade 3, with 2026-27 as the target year; Ministry replies since December 2025 say the end of Grade 2. It sets learning targets, designs materials and funds teacher training.
Vidya Pravesh
A three-month school preparation module proposed in NEP 2020 and launched in 2021. PRS reports 8.9 lakh schools implementing it as of December 2025 and 4.2 crore children covered in 2024-25.
Source: PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026. ASER 2024's rise in Std III reading and arithmetic is consistent with this focus, though a survey cannot attribute it to one programme.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series75www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Early years
Learning gaps open before school starts
Article 45 now asks the state to provide early childhood care and education until age six. NEP 2020 brings ages 3 to 5 into the foundational stage. GEEAP rates parent-directed stimulation for children aged 0 to 36 months and quality pre-primary education for ages 3 to 5 as good buys.
83.4%
of 4-year-olds in a pre-primary institution, 2024
ASER Centre, ASER 2024 National findings (rural India), released January 2025
16.7%
of Std I children underage (5 or below), lowest ever
ASER Centre, ASER 2024 National findings (rural India), released January 2025
Anganwadi centres, run under the women and child development ministry, enrol more than half of all 3 and 4 year olds (ASER 2024).
PRS reports 52% of pre-primary teachers professionally unqualified in 2024-25. Expanding pre-school without trained teachers risks pushing formal lessons onto four-year-olds.
See Child Development 101 for the developmental science.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series76www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Girls' education
Bangladesh's stipend for girls in secondary school
Bangladesh introduced the Female Secondary Stipend and Assistance Program (FSSAP) in 1994. Girls in rural secondary schools received a stipend and tuition subsidy if they attended 75% of school days, scored at least 45% in class tests and stayed unmarried until completing secondary school.
13% a year
growth in girls' secondary enrolment after 1994, against 2.5% for boys
ADB Development Asia policy brief on FSSAP
+1 year
rise in women's age at first marriage since the programme began
ADB Development Asia policy brief on FSSAP
3.14
estimated benefit-cost ratio
ADB Development Asia policy brief on FSSAP
Bangladesh now reports gender parity in primary and secondary education (Wikipedia, Education in Bangladesh). GEEAP lists targeting interventions towards girls as promising and merit scholarships for disadvantaged children as a good buy.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series77www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
From pilot to policy
Questions to ask before scaling an intervention
Most of the evidence in this section began as a pilot run by an NGO or researchers. Banerjee et al. (2017) show that results from small proof-of-concept studies may not hold when a government runs the programme. Muralidharan and Singh show a reform can be delivered on paper and change nothing.
QuestionWhy it matters
Who delivered the pilot?NGO staff are selected and supervised differently from government teachers
What did it cost per child per SD gained?Compare with the next best use of the same money
Does it change incentives or only add tasks?Tasks without incentives become paperwork
Did gains last after it stopped?Balsakhi gains fell from 0.28 to about 0.10 SD
Has it worked in more than one place?GEEAP's great buys have
See Cost Effectiveness 101 and Impact Evaluation 101 for the methods.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series78www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
10
Section Ten
Neighbours and higher education
CC BY-NC-ND 4.0 · ImpactMojo 101 Series79www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Spending in South Asia
India spends the most of its GDP among its large neighbours
Government expenditure on education as a share of GDP from the World Bank's World Development Indicators, which draws on UNESCO Institute for Statistics data. The latest available year differs by country, so read small differences with care.
Government expenditure on education, % of GDP (latest year)
World Bank, World Development Indicators (UNESCO UIS data), accessed 6 October 2026
Sri Lanka's share fell to 1.2% in 2022 during its economic crisis. Nepal's has fallen from 4.43% in 2021. Only India is above 4%, and none reaches the 6% that Indian policy has targeted since 1966.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series80www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Comparison table
Five South Asian systems on the same indicators
The table uses one source for every cell so the definitions match. Years are the latest available in WDI as of October 2026. Learning poverty is the World Bank's measure combining children out of school with those in school who cannot read an age-appropriate text by age 10.
IndicatorIndiaBangladeshPakistanNepalSri Lanka
Education spend, % of govt expenditure14.2 (2022)11.9 (2025)9.8 (2023)10.8 (2025)7.2 (2024)
Adult literacy, %78.2 (2024)79.0 (2022)58.9 (2021)68.7 (2019)92.7 (2024)
Primary completion rate, %90.3 (2025)94.3 (2024)74.1 (2024)109.4 (2025)94.5 (2023)
Tertiary GER, %34.4 (2025)23.7 (2024)10.9 (2024)21.3 (2024)20.7 (2023)
Learning poverty, %56.1 (2017)51.2 (2022)79.5 (2019)n/a14.8 (2015)
Source: World Bank, World Development Indicators (UNESCO UIS data), accessed 6 October 2026. Primary completion above 100 reflects over-age completers. Learning poverty years differ widely; use it for orders of magnitude only.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series81www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Pakistan
A constitutional right and 26 million children out of school
The Constitution (Eighteenth Amendment) Act 2010 inserted Article 25A into Pakistan's constitution: "The State shall provide free and compulsory education to all children of the age of five to sixteen years in such manner as may be determined by law." The amendment received presidential assent on 19 April 2010, eighteen days after India's Article 21A came into force.
26.2 million
children out of school, 2021-22
Pakistan Institute of Education, Pakistan Education Statistics 2021-22, via Geo News, 21 January 2024
39%
of school-age children out of school (44% in 2016-17)
Pakistan Institute of Education, via Geo News
65%
of children out of school in Balochistan
Pakistan Institute of Education, via Geo News
1.95%
of GDP on education, 2023
World Bank, World Development Indicators (UNESCO UIS data), accessed 6 October 2026
Pakistan's age band (5-16) is wider than India's (6-14), so its right covers secondary school. The share out of school fell while the number rose, because the school-age population grew faster.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series82www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Bangladesh
Gender parity on a small budget
Article 17 of the Constitution of Bangladesh commits the state to free and compulsory education for all children. Bangladesh is the region's standard example of reaching girls: the 1994 secondary stipend described in Section 9 coincided with rapid gains in girls' enrolment, and the country now reports gender parity in primary and secondary education.
2.03%
of GDP on education, 2024
World Bank, World Development Indicators (UNESCO UIS data), accessed 6 October 2026
79.0%
adult literacy, 2022
World Bank, World Development Indicators (UNESCO UIS data), accessed 6 October 2026
Spending has stayed close to 2% of GDP. Since 2011 WDI records a low of 1.1% (2019) and a high of 2.2% (2012).
Bangladesh shows that targeted demand-side incentives can move enrolment quickly even with a low budget. Learning is a separate question: WDI's learning poverty estimate is 51.2% for 2022.
Compare with Pakistan: similar spending as a share of GDP (1.95% in 2023), but primary completion of 74.1% against Bangladesh's 94.3% (WDI, 2024). The size of the budget does not settle results on its own.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series83www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Nepal
A federal constitution with education rights to secondary level
Article 31 of the Constitution of Nepal 2015 goes further than India's Article 21A. Clause (2) gives every citizen "the right to compulsory and free basic education, and free education up to the secondary level". Clause (5) gives every Nepali community the right to education in its mother tongue up to the secondary level.
Mother tongue as a right
Clause (5) makes mother-tongue education to secondary level a community right, with a right to open and run schools as provided by law. In India the same idea sits in NEP 2020 as a policy preference "wherever possible".
The money
WDI shows education spending of 3.69% of GDP in 2024, down from 4.43% in 2021, and 10.8% of government expenditure in 2025.
Compare Nepal's secondary-level guarantee with NEP 2020's 100% GER target for secondary by 2030: Nepal wrote it as a right, India as a goal.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series84www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Sri Lanka
Free education since 1945, and a fiscal squeeze
Sri Lanka introduced free education under C. W. W. Kannangara, the State Council's first Minister of Education. The reforms came into operation on 1 October 1945. Kannangara also built central schools in rural areas, from three in 1941 to 35 by 1945 and 50 by 1950. Undergraduate education at state universities remains tuition-free.
92.7%
adult literacy, 2024, the highest of the five countries here
World Bank, World Development Indicators (UNESCO UIS data), accessed 6 October 2026
14.8%
learning poverty, 2015
World Bank, World Development Indicators (UNESCO UIS data), accessed 6 October 2026
1.83%
of GDP on education, 2023 (1.2% in 2022)
World Bank, World Development Indicators (UNESCO UIS data), accessed 6 October 2026
Sri Lanka shows that early universal provision can produce lasting gains in literacy. It also shows how a fiscal crisis squeezes education: 7.2% of government expenditure in 2024 (WDI). History: Wikipedia, C. W. W. Kannangara and Education in Sri Lanka.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series85www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Higher education in India
Four crore students, uneven across states
Higher education in India is provided through institutions of national importance, central, state, private and deemed-to-be universities and their colleges. PRS reports 1,395 universities as of January 2026 and around four crore students enrolled.
1,395
universities, January 2026
PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026, citing AISHE dashboard
28%
higher education GER, 2021-22
PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026, citing AISHE 2021-22
50% by 2035
NEP 2020 target GER
National Education Policy 2020, Ministry of Education
Rs 55,727 cr
Department of Higher Education, 2026-27 BE
PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026
State gaps are wide: GER of 47% in Tamil Nadu and 41% in Kerala against 17% in Bihar, 19% in Jharkhand and 24% in Uttar Pradesh (AISHE 2021-22, via PRS).
WDI's tertiary GER for India is 34.4% for 2025, on UNESCO's definition, which differs from AISHE's. Do not mix the two series.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series86www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
One regulator
From HECI to the Viksit Bharat Shiksha Adhishthan Bill
NEP 2020 proposed a single Higher Education Commission of India (HECI) with four verticals: a regulatory council (NHERC), an accreditation council (NAC), a grants council (HEGC) and a General Education Council (GEC). The bill that followed took a different name and a slightly different shape.
The VBSA Bill 2025
Introduced in the Lok Sabha on 15 December 2025, it would create the Viksit Bharat Shiksha Adhishthan as a single regulator for higher education, replacing the UGC (set up under the UGC Act 1956), AICTE and NCTE, with regulatory, standards and accreditation councils. Section 22 of the UGC Act reserves degrees to universities set up by law, deemed universities and bodies empowered by an Act of Parliament; the Bill would let the Regulatory Council authorise others, with the Centre's approval. It went to a Joint Parliamentary Committee on 16 December 2025.
Status, as of October 2026
The committee listed its draft report for adoption on 17 July 2026; ThePrint (19 July 2026) reported that sitting called off. It was hearing experts again on 25 August and 11 September 2026, and PRS records no report presented. The Bill has not been passed.
Sources: PRS Legislative Research, Demand for Grants 2026-27 Analysis: Education, 17 February 2026; PRS bill and committee pages, accessed 6 October 2026; UGC Act 1956, s22. Check for a later report or vote before citing the status.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series87www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
11
Section Eleven
A practitioner's toolkit
CC BY-NC-ND 4.0 · ImpactMojo 101 Series88www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Diagnostic checklist
Ten questions before you design an education programme
Use these questions at the start of any programme design, district plan or funding proposal. Each one maps to a section of this course and to a data source you can open yourself.
  • Are children enrolled? (UDISE+, ASER)
  • Are they attending on a typical day? (your own spot checks)
  • Can Grade 3 children read a Grade 2 text? (ASER-style test)
  • Which children are furthest behind, and who are they?
  • Are teachers present and teaching? (unannounced visits)
  • How many posts are vacant, and in which schools?
  • Which scheme funds the activity, and is money released on time?
  • Does the school management committee meet and know the results?
  • Which state rules apply (detention, 12(1)(c), language)?
  • What will the programme cost per child, and per SD of learning?
If you can answer only the first question, you are designing an access programme. Most South Asian systems now need learning programmes.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series89www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Decision table
Match the problem to the evidence
The table links common field diagnoses to responses with the strongest evidence in this course, and to the trap that most often undoes them.
DiagnosisResponse with evidenceWatch out for
Many children in Grades 3-5 cannot readTeaching at the Right Level; structured pedagogyRunning it as a one-off camp with no follow-up
High teacher absenceRegular monitoring with consequencesMonitoring that adds forms without anyone acting on them
Parents cannot judge schoolsSimple learning report cardsNarrow tests that invite teaching to the test
Girls leave after Grade 8Conditional stipends; reducing travel timeConditions that exclude the poorest girls
Low pre-school qualityTrained staff and play-based materialsFormal lessons pushed down to age four
Unused 12(1)(c) seatsSeat publication, help with forms, prompt reimbursementOnline-only portals
CC BY-NC-ND 4.0 · ImpactMojo 101 Series90www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Worked example
Illustrative: planning a reading programme for one block
Illustrative. A block education officer has 120 government primary schools and about 6,000 children in Grades 3 to 5. A quick one-on-one test in a sample of schools suggests that roughly two-thirds cannot read a Grade 2 text. The officer has a modest budget from the state's foundational learning allocation.
01
TEST: every child in Grades 3-5, one-on-one
→
02
GROUP: by level for one hour a day
→
03
TRAIN: teachers in level-based activities
→
04
RETEST: at 50 days and at year end
What to measure
Share of children reading a Grade 2 text, by school and by group, at baseline and after each cycle. Teacher presence on unannounced visits.
What to cost
Training days, materials and supervision per child. Divide by the change in the share reading to get a cost per additional reader.
All numbers on this slide are illustrative. The design follows the TaRL evidence in Section 9.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series91www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Choosing a data source
Which dataset answers which question
Practitioners lose time arguing over numbers that come from instruments built for different purposes. This table pairs common questions with the source designed to answer them.
QuestionBest sourceLevel available
How many schools, teachers and students?UDISE+ (annual)School, block, district, state
Can children read and do arithmetic?ASER (household, rural)District and state
How do Grade 3, 6, 9 students score on the curriculum?PARAKH Rashtriya SarvekshanDistrict and state
Why are adolescents out of school?PLFS and NSO education surveysState
What does the Union spend, and is it used?Union budget, PRS analysisScheme
Higher education enrolment and institutions?AISHEState, institution
When sources disagree, check sample, year and method first. See Data Literacy 101 and Survey Design 101.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series92www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Working with government
Where an NGO or researcher can plug in
Most children are in government schools, so most lasting change runs through the state's own systems. Partnerships work best when they fit the calendar and the legal hooks the system already has.
System entry points
State annual work plans under Samagra Shiksha; district and block training calendars; state assessment cells and SCERTs; school management committees under RTE s21.
Legal hooks
s12(1)(c) admissions and their portals; s17 on corporal punishment; s24 teacher duties; the state's rules on examinations in classes 5 and 8.
Practical habits: share findings with the officials who supplied the data before publishing; present results by school and block, which is where decisions are made; and keep the intervention cheap enough that the state could run it without you.
See Governance & Accountability 101 for the wider accountability tools.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series93www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Children's data
Collecting learning data lawfully under the DPDP Act
Learning assessments collect personal data about children. The Digital Personal Data Protection Act 2023 and its 2025 Rules apply to it from 13 May 2027, when the duties, including section 9, commence (G.S.R. 843(E)). Section 2(f) defines a child as anyone who has not completed eighteen years.
Section 9
Before processing a child's personal data, a data fiduciary must obtain "verifiable consent of the parent" (s9(1)), and must not process data in a way "likely to cause any detrimental effect on the well-being of a child" (s9(2)).
Section 17(2)(b)
The Act's provisions do not apply to processing "necessary for research, archiving or statistical purposes" where the data are not used to take a decision about an individual and processing follows prescribed standards. It applies from 13 May 2027, with sections 9 and 11 to 17.
A test used to place a child in a learning group is a decision about that child, so do not assume the research exemption covers programme data. See Data Protection & the DPDP Act 101 and Research Ethics 101.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series94www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
12
Section Twelve
Open debates and where next
CC BY-NC-ND 4.0 · ImpactMojo 101 Series95www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Open debates
Six questions Indian education policy has not settled
Each of these has evidence on more than one side, and each is decided state by state as much as nationally. A practitioner should know the arguments before taking a position.
DebateOne sideThe other side
Detention in classes 5 and 8Exams restore effort and accountabilityHolding back without remediation pushes children out
Mother-tongue instructionChildren learn to read faster in a language they speakParents want English for jobs and mobility
School consolidationLarger schools get subject teachersLonger journeys keep the youngest and girls at home
Private school growthLower cost per child, similar resultsStratification and weaker public school constituencies
A single higher education regulatorLess duplication, clearer rulesCentral power over state and autonomous institutions
Contract teachersFaster hiring, local accountabilityLower pay, weaker careers, legal challenges
CC BY-NC-ND 4.0 · ImpactMojo 101 Series96www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
What to watch
Dates and decisions to watch after October 2026
Education policy changes through survey releases, budgets, court orders and bills. These are the scheduled or pending events, as of October 2026, that are most likely to change the numbers and rules in this deck.
  • ASER 2026: Pratham is preparing the next survey; ASER 2024 remains the latest report
  • UDISE+ 2026-27: the next annual school census
  • NIPUN Bharat: 2026-27 is the mission's target year for foundational learning
  • Census 2027: reference date 1 March 2027, including caste enumeration, will reset enrolment denominators
  • Pramati reference: a larger bench on whether the RTE Act applies to minority schools
  • Dinesh Biwaji Ashtikar: hearing on 27 October 2026 on 12(1)(c) rules
  • VBSA Bill: JPC report and any vote in Parliament
  • Union Budget 2027-28: the next test of the 6% target
Re-check each item before you cite it. Dates move, and court matters are often adjourned.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series97www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Summary
Eight ideas to take away
  • Enrolment is near universal for ages 6-14; learning is the binding constraint
  • Article 21A and the RTE Act made elementary education a justiciable right from 1 April 2010
  • Education is on the Concurrent List, so states decide much of what NEP 2020 becomes
  • The 6% of GDP target dates from 1966; spending was 4.1% in 2022-23
  • Teacher absence and vacancies cost more than most new schemes add
  • Private schools now teach 38% of students; selection explains much of their raw advantage
  • Great buys: information, structured pedagogy, teaching at the right level
  • Pilots need to survive government delivery before they count as policy
These severe shortfalls constitute a learning crisis.
World Bank, World Development Report 2018, Overview
The figures behind each idea, with sources, are in Sections 1 to 10. When you quote one, quote its year with it, because most of them will change within two years.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series98www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Where next
Related ImpactMojo 101 decks
Education policy connects to almost every other part of development practice. These decks go deeper on the methods, rights and systems this course touched.
Item Response Theory 101 explains how assessments such as PARAKH put different test booklets on one scale.
CC BY-NC-ND 4.0 · ImpactMojo 101 Series99www.impactmojo.in
ImpactMojoEducation Policy 101www.impactmojo.in
Education Policy 101 · Complete
Count the children,
then count what they learn
Enrolment was the twentieth-century problem. Learning, teachers and money spent well are this century's. Read the law, open the data, test before you scale, and keep asking which child is furthest behind. Explore the rest of the ImpactMojo 101 Series, free forever.
More 101 CoursesExplore ImpactMojoDataverse
CC BY-NC-ND 4.0·Free Forever·ImpactMojo 101 Series
CC BY-NC-ND 4.0 · ImpactMojo 101 Series100www.impactmojo.in