Skip to content
MangoPOV

Complex things, explained from the documents.

MangoPOV takes one Indian institution, law or number at a time and explains it from the record that created it. Every figure you read carries the document it came from, and that document is held here in full.

91
Documents held
47
Government primary sources
0
Figures with no source

ReservationReservation in IndiaPart 11 · 11 published

EWS eligibility explained: the Rs 8 lakh limit, and who cannot apply

The Constitution created a reservation for the economically weaker and put no number in it, leaving the line to be notified by the State from time to time. A departmental memorandum set it at Rs 8 lakh a year, with asset bars that exclude a family irrespective of its income. The groups holding most of India's poorest cannot apply at all, because they are covered somewhere else.

Rs 8 lakh
Gross annual family income below which a household qualifies as economically weaker
Rs 47 a day
What the same government's poverty index treats as poor in a city, and Rs 32 in a village
48.4%
Of Scheduled Tribes below the poverty line on the Sinho Commission's figures, and they cannot apply
82%
Of the country's population in the groups the quota excludes by definition rather than by income

15 September 2026 · 9 min read · 4 documents

Read the piece

More to read

  1. Part 10 · Reservation in India

    EWS reservation explained: the 103rd Amendment, and the 50% ceiling

    In 1992 nine judges fixed reservation at 50 per cent and held that economic criteria alone cannot decide who is backward. In 2019 Parliament created a 10 per cent quota on economic criteria alone, on top of the 50. In 2022 five judges upheld it, three to two, not by moving the ceiling but by holding it had never applied to this clause. The Chief Justice was in the minority.

    59.50%
    Total reservation once the 10% for economically weaker sections is added to 15%, 7.5% and 27%
    68 per cent
    What Balaji actually struck down in 1962, in a judgment that declined to fix an outer limit
    54 judgments
    What counsel counted over 60 years, all saying the ceiling must be maintained
    3 to 2
    The majority upholding the amendment, with the Chief Justice signing the minority opinion

    15 September 2026 · 9 min read · 2 documents

  2. Part 9 · Reservation in India

    Jarnail Singh and B.K. Pavitra explained: the data test for promotion quotas

    A Karnataka law on promotions fell in 2017 because the State had never counted what it was required to count. It constituted a committee and had a report in forty-four days, and won on it. Five years later the Supreme Court held the counting unit itself was the wrong one, and the Union of India, asked about its own establishment, could answer for 44 of about 90 Ministries.

    44 days
    From the Ratna Prabha Committee's constitution to its report on backwardness, representation and efficiency
    10.65% / 2.92%
    Scheduled Caste and Scheduled Tribe representation across 31 departments, measured against sanctioned posts
    80.35%
    Of those sanctioned posts sitting in six departments: Education, Home, Health, Revenue, Judicial and Finance
    44 of 90
    Ministries and Departments that gave the Union of India cadre information, covering 30 lakh employees

    15 September 2026 · 11 min read · 3 documents

  3. Part 8 · Reservation in India

    Reservation in promotion explained: Article 16(4A), and the M. Nagaraj case

    In 1992 the Supreme Court held that reservation stops at the door of a government job and cannot climb the ladder inside it. Parliament wrote the ladder back in four times in six years, once by a clause that took effect six and a half years before it was signed. In 2006 five judges upheld every one of them, and replaced the words in the opinion of the State with a demand for numbers.

    49.5%
    Reservation already reached in a year when clause (4B) took the backlog out of the ceiling
    17 June 1995
    The date the Eighty-fifth Amendment is deemed to have come into force, having been assented on 4 January 2002
    4 amendments
    The 77th, 81st, 82nd and 85th, all upheld by five judges in one sentence
    quantifiable data
    What a State must now collect on backwardness and inadequate representation before it reserves in promotion

    15 September 2026 · 10 min read · 9 documents

  4. Part 7 · Reservation in India

    Indra Sawhney case explained: the 50% ceiling, and the creamy layer

    Nine judges upheld the 27% quota, struck down the 10% beside it, and fixed the ceiling at 50 per cent over the objection of some of their own number. In the same judgment they held that no reservation can be made on economic criteria, and ordered an income test to decide who should be excluded.

    49.5%
    Total reservation under the memoranda, half a point under the ceiling
    a year
    The unit for applying the 50% rule, not the strength of the cadre
    5 years
    The runway given to existing promotion reservations before the bar applied
    2,342
    Entries in the Central List of Other Backward Classes, and 2,449 in the State Lists

    14 September 2026 · 9 min read · 2 documents

  5. Part 6 · Reservation in India

    Mandal Commission explained: how 11 indicators produced the 27% OBC quota

    A caste scoring 11 points out of 22 was a backward class and a caste scoring 10 was not. The test found 3,743 castes and put them at 52 per cent of India. The Commission then recommended 27 per cent, saying in its own report that it did so in deference to legal limitation, which left the total half a percentage point under the courts' ceiling.

    11 of 22
    The score that made a caste a backward class; 10 did not
    3,743
    Castes identified, against 2,399 by the first commission and 1,057 by another survey
    12.55%
    Of government employment held by a group put at 52% of the population, and 4.69% of Class I
    27%
    Recommended in deference to legal limitation, not because the findings supported it

    14 September 2026 · 9 min read · 2 documents

  6. Part 5 · Reservation in India

    Kalelkar Commission explained: how India first tried to define who is backward

    The Constitution let the State make special provision for backward classes and never said what one was. Ambedkar's answer in the Assembly was that a backward community is one that is backward in the opinion of the government. The first commission's own chairman asked for his report to be rejected, and the 50% ceiling everyone now argues about began as a sentence that said "speaking generally and in a broad way".

    94.61%
    Of Tamil Nadu's Muslim population inside the backward list
    83 posts
    District Munsif vacancies split across six communal columns in 1949
    less than 50%
    The whole of the ceiling, as Balaji actually worded it
    65%
    What a 17.5% quota reached in its third year under the carry-forward rule

    14 September 2026 · 10 min read · 3 documents

  7. Part 4 · Reservation in India

    Article 334 explained: how ten years of reserved seats became eighty

    Article 334 reserved seats for the Scheduled Castes and Scheduled Tribes and set them to expire on 26 January 1960. Six amendments then bought exactly 10 more years each. The seventh, in 2019, gave the reserved seats 80 years and gave the Anglo-Indian nominated seats 70, a term that had already been served.

    1960
    The date Article 334 originally set for reservation of seats to expire
    80 years
    The period now written in, running to 2030
    1971
    The census entitlement is pinned to, by Act 108 of 1976
    6 of 14
    Seats to Non-Brahmin Hindus under the Madras grid struck down in 1951

    14 September 2026 · 9 min read · 4 documents

  8. Part 3 · Reservation in India

    Constituent Assembly 1949: the tribal census dispute behind reserved seats

    Seats were reserved in proportion to population. In the same debate, the member speaking for India's tribal communities told the House the census had been cooked at every count, and in Assam 8 lakh tribals in the tea gardens were counted into the population and out of the reserved seats.

    24.8m
    Adivasi population in the 1941 census, against 30 million claimed by scientists
    24 of 1,585
    Adivasi members in the provincial legislatures under the 1935 Act
    8 lakh
    Assam tea-garden tribals, counted in the population and out of the reserved seats
    18 lakh
    The Central Provinces Adivasi figure alleged lost to misenumeration, 1911 to 1941

    14 September 2026 · 10 min read · 4 documents

  9. Part 2 · Reservation in India

    Constituent Assembly 1949: why India's minorities gave up reservation

    In the Constituent Assembly's Minorities Committee, the Muslim, Christian and Sikh communities agreed to surrender reservation of seats. On Sardar Patel's account the amendment to end it was moved by a Muslim member, and the committee split 5 to 2 on one member's own uncertain count. What survived was reservation for the Scheduled Castes alone, with 10 years written beside it.

    33
    Muslim members of the Assembly, in a debate about protecting a minority
    5 to 2
    The committee split, on the mover's own uncertain count
    12.5% and 16.5%
    Set aside for Scheduled Castes in competitive and non-competitive services, by executive order
    10 years
    The limit Patel accepted on the seats that survived

    14 September 2026 · 12 min read · 3 documents

  10. Part 1 · Reservation in India

    Poona Pact explained: what Ambedkar traded, and the clause that never expired

    The Poona Pact set an expiry date on one of the things it created and none on the other. The primary elections were to end after ten years; the reserved seats were to continue until the communities concerned agreed otherwise. No such agreement is on record, and the clause is now in its 94th year.

    148
    Seats reserved from the general electorate, 78 more than the Communal Award offered
    18%
    Of general-electorate seats in the Central Legislature, written into the text
    10 years
    The only expiry in the Pact, and it covered the primary elections, not the seats
    1952
    When a member told the Constituent Assembly reservation should have ended

    14 September 2026 · 9 min read · 5 documents

  11. Reservation

    Maratha reservation explained: the backward class claim, and the 50% ceiling

    A community that has given Maharashtra thirteen of its nineteen chief ministers has spent forty-six years asking to be classified as backward. Its own government's commission found 85 per cent of Maratha families living on under Rs 25,000 a year. Both claims come from the same court judgments.

    13 of 19
    Maharashtra chief ministers from the Maratha community
    76.86%
    Maratha families in agriculture or farm labour, per the state commission
    37.28%
    Below the poverty line, against a state average of 24.2%
    50%
    The reservation ceiling the quota crossed, fixed in 1992

    11 September 2026 · 16 min read · 23 documents

  12. Part 5 · The National Testing Agency

    NTA reforms after the NEET leak: the Public Examinations Act, 16 new posts, and NEET-UG 2026 cancelled

    After the NEET-UG 2024 leak the National Testing Agency was rebuilt: 16 new posts, 101 committee recommendations, and the Public Examinations Act 2024, a law written before the leak it is now measured against. Then NEET-UG 2026 was cancelled and a CBI case followed.

    16
    New posts created inside the agency
    10 years
    Maximum sentence for organised crime under the Act
    3 May 2026
    The exam that was held, then cancelled

    8 September 2026 · 10 min read · 8 documents

  13. Part 4 · The National Testing Agency

    UGC-NET 2024 cancelled in a day, and the Radhakrishnan committee's 101 recommendations

    UGC-NET was held on 18 June 2024 and cancelled on 19 June. Sixty-seven candidates had scored a perfect 720 in NEET-UG. An expert committee was given two months, took four, and returned 101 recommendations.

    67
    Candidates with a perfect 720 out of 720
    1 day
    Between UGC-NET being held and being cancelled
    101
    Recommendations the expert committee proposed

    8 September 2026 · 10 min read · 15 documents

  14. Part 3 · The National Testing Agency

    NEET-UG 2024 paper leak: Hazaribagh to Patna, and why the Supreme Court refused a re-exam

    The NEET-UG 2024 paper leak, from a raid on the morning of the exam to burnt question-paper fragments matched to a centre in another state. The Supreme Court called the leak an admitted fact, and let the examination stand anyway.

    23,33,297
    Candidates who sat the paper on 5 May 2024
    155
    Students the CBI material indicated appear to be beneficiaries
    2
    Centres where the leak is not in dispute

    8 September 2026 · 16 min read · 13 documents

  15. Part 2 · The National Testing Agency

    NEET-UG 2024 grace marks: 44 perfect scores, 1,563 scorecards cancelled, and a re-test

    Grace marks carried 44 candidates in NEET-UG 2024 to a full score. Inside the same affected group, scores ran from minus 20 to 720. The scorecards of 1,563 candidates were then cancelled and a re-test ordered, and a separate CBI case from JEE Main 2021 remains, in the agency's own word, continuing.

    44
    Candidates reached a full 720 after grace marks
    −20 to 720
    The score range inside the same affected group
    1,563
    Scorecards cancelled, with a re-test offered

    8 September 2026 · 12 min read · 8 documents

  16. Part 1 · The National Testing Agency

    NTA explained: why the National Testing Agency is a society, not a statutory body

    The National Testing Agency was not created by an Act of Parliament. It was registered as a society under a law from 1860, with one opening grant of Rs 25 crore and an instruction to pay for itself thereafter.

    1860
    Year of the law it is registered under
    Rs 25 cr
    Granted once, then self-sustaining
    40 lakh
    Students the estimate covers

    8 September 2026 · 12 min read · 9 documents

Who the documents come from

See all 91 documents

Primary sources

  • National Testing Agency9
  • Supreme Court of India8
  • Indian Kanoon6
  • Constituent Assembly of India5
  • Supreme Court of India, via Indian Kanoon4
  • Press Information Bureau, Government of India3
  • Department of Personnel and Training, Ministry of Personnel, Public Grievances and Pensions2
  • Legislative Department, Ministry of Law and Justice2
  • Act 1 of 20241
  • Backward Classes Welfare Department, Government of West Bengal1
  • Bombay High Court, via Indian Kanoon1
  • Central Bureau of Investigation1
  • Gazette of India1
  • Government of India1
  • Journal of Indian Education, NCERT1
  • National Commission for Backward Classes1

A gazette, a court order, a Cabinet decision, or a body's own filing.

Reporting

  • DD News14
  • Akashvani News8
  • Deccan Herald7
  • Down To Earth2
  • Free Press Journal2
  • PRS Legislative Research2
  • Scroll.in2
  • The Quint2
  • Centre for Law and Policy Research, Constitution of India1
  • Drishti IAS1
  • Supreme Court Observer1
  • The Federal1
  • Wikipedia1

Used where the primary was sought and this is what exists. Labelled as reporting on every piece it appears in.

What you get here

Traceable figures
Every number carries a mark you can point at, and it names the document behind it, who published it, and the date this archive fetched its bytes.
Ranked sources, shown
A gazette, a court order or an answer in Parliament counts as a primary source. Anything weaker is labelled as weaker, on the page, rather than quietly.
Alleged stays alleged
Nothing under investigation or before a court is written as settled until a court concludes it. Private individuals are not named.

How a claim gets onto the page

These pieces are written over an archive rather than from notes. The steps are mechanical, which is the point: they do not depend on anyone remembering to be careful.

  1. 01

    Fetched and stored

    The document is downloaded and its bytes kept, with a SHA-256 and the date. Indian government links rot; the copy a claim was checked against does not.

  2. 02

    Located, not paraphrased

    The exact wording behind a claim has to be found in those stored bytes. A claim whose wording cannot be located is not used.

  3. 03

    Ranked

    The document is classified by what kind of source it is, and that ranking is published beside it rather than kept internal.

  4. 04

    Checked before it ships

    Every figure is matched back to an approved claim. A number with nothing behind it does not reach the page.

What this checks is the sourcing, not the truth. A document can be primary and still record something unproven, which is why status and sourcing are shown as two separate things. More on the method.