The HDR family of indices and the liberty critique

Detail

1. Why the HDI needs companion indices

• HDI (Human Development Index) is the geometric mean of three dimension indices: health, education and income (standard of living).

• Formula: HDI = (I_health × I_education × I_income)^(1/3).

• The HDI is a national average. It says nothing about:

• how the gains are shared among people (so UNDP built the IHDI),
• how they are shared between women and men (GDI, GII),
• the damage to the planet caused in reaching them (PHDI),

• overlapping deprivations in poor households (MPI).

• India's headline figure: HDI 0.685 (2023 data, HDR 2025), up from 0.676 (2022). Rank 130 of 193. India is still in the "medium human development" group, close to the 0.700 mark for "high" [2].

• India's HDI has risen by over 53% since 1990, which is faster than the world and South Asian averages [2].
• HDR 2025 is titled "A Matter of Choice: People and Possibilities in the Age of AI" [2].

2. The companion indices at a glance (UNDP)

Index	Since	What it does	India	Direction
Inequality-adjusted HDI (IHDI)	2010	Cuts each HDI dimension down by its Atkinson inequality measure. The gap between HDI and IHDI is the "loss due to inequality"	~0.47; loss 30.7% (HDR 2025) [2] (NCERT scaffold: loss ~30%)	IHDI ≤ HDI. It equals the HDI only under perfect equality
Gender Development Index (GDI)	1995 (revised 2014)	Female HDI ÷ male HDI. Countries fall into 5 groups by their absolute distance from parity	~0.87	Parity = 1
Gender Inequality Index (GII)	2010	3 dimensions, 5 indicators	~0.40, rank ~102 (HDR 2025)	0 = equality; higher = worse
Planetary pressures-adjusted HDI (PHDI)	2020	Adjusts the HDI for CO₂ emissions and material footprint per person (see Section 10 of the parent note)	—	PHDI ≤ HDI
Multidimensional Poverty Index (MPI)	2010 (with OPHI, Oxford)	Counts overlapping deprivations in health, education and living standards (covered in the poverty-inequality note)	—	—

3. Inequality-adjusted HDI (IHDI)

• Meaning: the IHDI is the HDI that people actually get, given how unequally it is shared. The HDI is the "potential" level, i.e. what they would get if everything were shared equally.
• Atkinson inequality measure (A): this measures how far the geometric mean of a distribution falls below its arithmetic mean.
• A = 1 − (geometric mean ÷ arithmetic mean).

• Worked example: two people earn 1 and 4. Arithmetic mean = 2.5. Geometric mean = √(1×4) = 2. So A = 1 − 2/2.5 = 0.20, meaning a 20% inequality discount.

• How the IHDI is built:

• Each dimension index is multiplied by (1 − A) for that dimension.

• IHDI = [(1−A_health)·I_health × (1−A_edu)·I_edu × (1−A_income)·I_income]^(1/3).

• Loss due to inequality (%) = (1 − IHDI/HDI) × 100.

• India check: 0.685 × (1 − 0.307) ≈ 0.475, which matches the IHDI of ~0.47 [2].

• India's pattern (HDR 2025):

• India's loss of 30.7% is one of the highest in South Asia [2].
• Inequality in health and education has come down, but income and gender gaps are still large [2].

4. Gender Development Index (GDI)

• Definition: GDI = HDI (female) ÷ HDI (male). It compares achievement levels of women and men.
• Worked example: female HDI 0.62 and male HDI 0.71 give GDI = 0.62/0.71 ≈ 0.87. That is India's approximate level.
• The 5 groups: countries are grouped by the absolute distance of their GDI from 1.
• Group 1 is closest to parity. Group 5 is farthest (more than 10% away).
• India's gap is about 13% (1 − 0.87), which puts it in the farthest group.

• Distance counts in either direction. A GDI above 1 (women ahead of men) is also a gap from parity.

• Revised in 2014: the old GDI (1995) was a gender-sensitive HDI. The revised version is a simple female-to-male ratio.

5. Gender Inequality Index (GII)

• Definition: the GII measures the loss in achievement caused by gender inequality. It runs from 0 (full equality) to 1. Higher = worse.
• 3 dimensions, 5 indicators:
• Reproductive health (measured for women only):
  • maternal mortality ratio (MMR): mothers who die from pregnancy-related causes per 1 lakh live births;
  • adolescent birth rate: births per 1,000 women aged 15–19.
• Empowerment:
  • share of parliamentary seats held by each sex;
  • population with at least some secondary education (female vs male).

• Labour market:

  • labour force participation rate (LFPR): the share of working-age people who are working or looking for work (female vs male).

• India's GII value: about 0.40, rank about 102 (HDR 2025).

• Indian data behind the indicators:
• MMR fell from 130 (2014–16) to 93 (2019–21) per lakh live births (Sample Registration System) [5].
• Female LFPR (age 15+, usual status) rose from 37.0% (2022–23) to 41.7% (2023–24) (PLFS) [6].
• Parliament: 75 women were elected to the Lok Sabha in 2024, about 14% of 543 seats [3][4].
• Constitution (106th Amendment) Act, 2023, the Nari Shakti Vandan Adhiniyam, reserves one-third of seats for women in the Lok Sabha and the State Legislative Assemblies, including Delhi's, on a rotating basis [3]. UNDP says this "offers promise" for the empowerment dimension [2].
• UNDP says female LFPR and political representation are where India lags most [2].

6. The GDI–GII trap

• In the GDI, a value near 1 is good. It is a ratio, and parity is 1.
• In the GII, a higher value means more inequality. Here 0 is ideal.
• The GDI compares achievements (HDI of women vs men). The GII measures disadvantage, including reproductive health, which applies to women only.
• Do not read the two indices the same way.

7. Planetary pressures-adjusted HDI (PHDI) and MPI (links only)

• PHDI (2020): HDI multiplied by an adjustment factor for CO₂ emissions per person and material footprint per person. The more a country loads the planet, the further its PHDI falls below its HDI. Details are in Section 10 of the parent note.
• MPI (2010, UNDP with OPHI): measures poverty through overlapping deprivations, not income. Details are in the poverty-inequality note.

8. The liberty critique (Class 11, Comparative Development Experiences, §8.5)

• NCERT's position: the HDR indicators are "extremely important … but these are not sufficient". We also need liberty indicators.
• What was added: one measure, "the extent of democratic participation in social and political decision-making". It was "not given any extra weight".
• Liberty indicators that are missing:
• the extent of constitutional protection of citizens' rights;

• the extent of constitutional protection of the independence of the judiciary and the rule of law.

• NCERT's verdict: unless these are included and "given overriding importance", the HDI is "incomplete and its usefulness limited".

• Why it matters for India vs China:
• China is ahead of India on health and education.
• But in China, "lack of political freedom and its implications for human rights are major concerns" (NCERT conclusion).

• So a country can score well on HDI while its people have little freedom. The HDI alone cannot show this.

• UNDP's own attempts:

• Human Freedom Index (HDR 1991) and Political Freedom Index (HDR 1992).
• Both were dropped after member states objected.

• This shows how politically hard it is to measure freedom inside an organisation of governments.

• Link to the capability approach: the HDI comes from Amartya Sen's idea of development as expanding people's freedoms and choices. Leaving out political freedom is therefore a gap in the index's own foundation.

9. Other critiques of the HDI

• Arbitrary equal weights. Each dimension gets one-third, and no theory says why.
• Substitutability. A gain in one dimension can make up for a loss in another. The geometric mean (used since 2010) reduces this but does not remove it.

• Worked example: health 0.9, education 0.3, income 0.9.

  • Arithmetic mean = 0.70.
  • Geometric mean = (0.9 × 0.3 × 0.9)^(1/3) = (0.243)^(1/3) ≈ 0.62.
  • The weak schooling pulls the score down more under the geometric mean, but a big income gain can still hide it.

• Years of schooling ≠ learning. The HDI counts years in school, not what children learn. In India, ASER surveys show learning outcomes lagging behind enrolment.

• Data lags and revisions.
• HDR 2025 uses 2023 data [2].
• UNDP recalculates past values with new data (back-calculation). So ranks from different reports cannot be compared. Compare only within the trend table of the same report.

• Example: India's rank was 132 in HDR 2023/24 and 130 in HDR 2025. Part of such a change can come from data revisions, not real progress.

• It hides inequality because it averages across a whole country. This is why the IHDI exists.

• It leaves out environment, security and freedom. The PHDI covers the environment. Freedom is still not covered (see Section 8).

10. Precursor: Physical Quality of Life Index (PQLI)

• Designed by Morris David Morris in 1979 for the Overseas Development Council (USA).
• Three indicators:
• infant mortality rate;
• life expectancy at age one (not at birth, so it does not double-count infant deaths);

• basic literacy rate.

• Each indicator is scaled 0–100 (worst = 0, best = 100). The three are then simply averaged (arithmetic mean).

• Worked example: scaled scores of 60, 70 and 80 give PQLI = (60+70+80)/3 = 70.

• No income component. It was built as an alternative to GNP, to show that welfare can differ from income.

• HDI vs PQLI: the HDI (1990) added income, and since 2010 it uses a geometric mean instead of a simple average.

Prelims Hooks

• IHDI (2010) discounts each HDI dimension by the Atkinson inequality measure. India's loss due to inequality is 30.7% (HDR 2025) [2].
• GDI = female HDI ÷ male HDI. 1 = parity, and countries fall into 5 groups by absolute distance from 1. A GDI above 1 is also a gap.
• GII: 0 = equality, higher = worse. It has 3 dimensions (reproductive health, empowerment, labour market) and 5 indicators: MMR, adolescent birth rate, parliamentary seats, secondary education, LFPR.
• Trap: MMR and adolescent birth rate are measured for women only. The other GII indicators compare women with men.
• India, HDR 2025: HDI 0.685 (2023 data), rank 130/193, medium human development [2].
• PQLI (Morris D. Morris, 1979): infant mortality, life expectancy at age one and literacy, simply averaged, with no income.
• UNDP's Human Freedom Index (1991) and Political Freedom Index (1992) were dropped after objections from member states.
• PHDI (2020) adjusts the HDI for CO₂ emissions and material footprint per person. MPI (2010) was built with OPHI.
• 106th Constitutional Amendment (2023), the Nari Shakti Vandan Adhiniyam, reserves one-third of Lok Sabha and State Assembly seats for women [3]. This bears on the GII's "parliamentary seats" indicator.

Mains Points

• The HDI is necessary but not sufficient (GS-II/III):
• NCERT's liberty critique shows that the HDI leaves out rights, judicial independence and the rule of law.
• The India–China comparison shows that a higher HDI can exist alongside weak political freedom.

• UNDP dropping its freedom indices (1991, 1992) shows the political limits of measuring freedom.

• Growth without equal sharing:

• India's HDI has risen by more than 53% since 1990 [2].
• But its inequality loss of 30.7% [2] and low female LFPR (41.7%, 2023–24) [6] show that gains are unevenly shared.

• Policy answer: invest in human capital (learning quality, not just enrolment), jobs for women, and targeted welfare.

• Gender as a development bottleneck:

• MMR fell to 93 (2019–21) [5], but women's LFPR and representation (75 women MPs, 2024) [3][4] keep the GII high at ~0.40.

• Once implemented, the 106th Amendment could improve the empowerment dimension [2][3].

• Choosing a measure is a value judgment:

• Equal weights, substitutability, schooling-versus-learning and data revisions all mean that rank changes should be read with care.
• A good answer uses a set of indices (IHDI, GII, PHDI, MPI) rather than a single HDI rank.