Human Development and Human Capital: HDI, IHDI, GDI, GII and India's National MPI, Precisely Distinguished

This chapter has produced exactly 3 real Prelims questions so far (2018 and 2019), on social capital versus human and physical capital, human capital formation through health and education spending, and the actual mandate of the Pradhan Mantri Kaushal Vikas Yojana. That is a thin PYQ record for a conceptually dense topic, and exactly the kind of gap UPSC tends to close rather than leave alone: five distinct UNDP indices that sound alike and measure different things, a national poverty index India built for itself, and a demographic argument that cuts both ways depending on whether human capital investment keeps pace.

This chapter's material overlaps with NCERT Class 12, Indian Economic Development, Chapter 5 (Human Capital Formation in India). This note goes further than that chapter's coverage.

The Human Development Index: three dimensions, one number

The Human Development Index (HDI) was developed by the United Nations Development Programme (UNDP), first published in the 1990 Human Development Report, built on the foundational work of economists Mahbub ul Haq and Amartya Sen. Its entire premise is a correction: national income alone is a poor proxy for whether people are actually living longer, more educated, better-resourced lives, so the HDI folds income together with health and education into one composite number.

The HDI rests on exactly three dimensions, each with a specific, UPSC-testable indicator:

• A long and healthy life, measured by life expectancy at birth.
• Knowledge, measured by a combination of expected years of schooling (for a child starting school today) and mean years of schooling (the average actually completed by adults aged 25 and above).
• A decent standard of living, measured by Gross National Income (GNI) per capita at purchasing power parity (PPP), not GDP per capita. UNDP's own technical notes are explicit that this is GNI, not GDP, since GNI additionally captures net income Indian residents and firms earn abroad.

The exam trap here is almost always the education dimension: it is not literacy or enrolment, and it is not a single number, it is the average of two distinct schooling measures, one prospective (expected years) and one retrospective (mean years actually completed).

The formula: from arithmetic mean to geometric mean, a real methodological change

Each dimension is first converted into a dimension index between 0 and 1, using fixed minimum and maximum "goalposts": life expectancy is normalised against a floor of 20 years and a ceiling of 85 years, expected years of schooling against 0 to 18 years, mean years of schooling against 0 to 15 years, and GNI per capita against $100 to $75,000 (in 2021 PPP dollars). The formula for each dimension index is:

Dimension index = (actual value − minimum value) ÷ (maximum value − minimum value)

This is genuinely worth knowing precisely, because the HDI has been revised more than once, and the single most exam-relevant change is this: until the 2010 Human Development Report, the HDI was the arithmetic mean (simple average) of the three dimension indices; from 2010 onward, UNDP switched to the geometric mean. The change was deliberate, not cosmetic. Under an arithmetic mean, a country could compensate for a very weak dimension (say, health) with a very strong one (say, income), because the three dimensions behaved as perfect substitutes for each other. The geometric mean removes that substitutability: HDI = (Health index × Education index × Income index)^(1/3). As any single dimension index approaches zero, the whole HDI is pulled toward zero regardless of how strong the other two dimensions are, so a country cannot buy its way to a high HDI through income alone while neglecting health or education. This is precisely the kind of "when was X changed and why" fact UPSC likes to test, and candidates who only remember "HDI averages three dimensions" without the 2010 arithmetic-to-geometric shift will get an option like this wrong.

The 2014 Human Development Report additionally fixed the cutoff points that sort countries into four human development categories: very high (0.800 and above), high (0.700 to 0.799), medium (0.550 to 0.699), and low (below 0.550), and the current Report keeps these same cutoffs.

India's HDI: the current numbers, precisely

The most recent Human Development Report is HDR 2025, titled "A Matter of Choice: People and Possibilities in the Age of AI," released by UNDP on 6 May 2025, reporting 2023 data (the HDI always lags by roughly two years, since it depends on internationally reconciled data). India's figures, straight from UNDP's own India press release:

India sits in the medium human development category (0.550 to 0.699), just below the 0.700 threshold for "high." Its HDI value has risen by over 53% since 1990, faster than both the global and South Asian averages, and its 2023 life expectancy of 72 years is the highest recorded since the index began for India, a genuine post-pandemic recovery signal. Because the Human Development Report Office revises the entire historical time series with every new release (updated data, updated goalposts), a rank or value from an older report should never be quoted as current; always anchor to the latest Report's own numbers, not a remembered figure from a prior year's edition.

Beyond HDI: IHDI, GDI and GII, the exam's favourite mix-up

UNDP publishes four related composite indices alongside the HDI, and UPSC's most reliable trap in this chapter is testing whether a candidate can tell them apart. All four share the same three basic dimensions (health, education, standard of living, adjusted for gender or inequality as appropriate), but they are not interchangeable.

Inequality-adjusted HDI (IHDI): discounts the HDI for internal inequality

The IHDI answers a different question from the HDI: not "how well is the average person doing," but "how well is human development actually distributed within the country." It is calculated by first adjusting each of the three dimension indices downward for the inequality observed within that dimension's distribution across the population (using an Atkinson-family inequality measure), and then taking the geometric mean of these three inequality-adjusted indices. The IHDI equals the HDI when there is no inequality, and falls below the HDI as inequality rises, so the gap between a country's HDI and its IHDI is itself a usable statistic: UNDP calls it the "loss" in human development due to inequality. For India, inequality reduces the HDI by 30.7%, one of the highest such losses in the region, a direct measure of how uneven India's health, education and income gains are within its own population, even as the headline HDI keeps improving.

Gender Development Index (GDI): female HDI against male HDI

The GDI is simply the ratio of the female HDI value to the male HDI value, each calculated using the identical three HDI dimensions and formula, just computed separately for each sex. A GDI close to 1 means women and men are achieving similarly across health, education and income; a GDI well below 1 flags a systematic gender gap in human development outcomes. The GDI is fundamentally a comparison of achievement levels, not a measure of empowerment or opportunity as such.

Gender Inequality Index (GII): a genuinely different index, not GDI's twin

This is where most candidates slip: GII is not another name for GDI. Where GDI compares two HDI values, the GII measures something GDI does not touch at all, gender-based disadvantage across three entirely different dimensions:

• Reproductive health, via the maternal mortality ratio and the adolescent birth rate.
• Empowerment, via the female and male shares of parliamentary seats and the female and male population with at least secondary education.
• Labour market, via female and male labour force participation rates.

A higher GII value indicates greater gender-based disadvantage (0 means women and men fare equally; 1 means women fare as poorly as possible on every measure), which is the opposite direction from most other UNDP indices, where higher is better, another common source of scoring-direction errors in MCQs. Remember the anchor: GDI = ratio of two HDI scores; GII = reproductive health + empowerment + labour market disadvantage, an index with its own indicators entirely.

Measuring poverty directly: the Multidimensional Poverty Index

Income-based poverty lines capture only one dimension of deprivation. The Multidimensional Poverty Index (MPI) was introduced in the 2010 Human Development Report and, since 2018, has been jointly produced by UNDP's Human Development Report Office and Oxford's Poverty and Human Development Initiative (OPHI). It identifies a person as multidimensionally poor if deprived across at least one-third of a weighted basket of indicators spanning health, education and standard of living. Two components combine to form the MPI value: the headcount ratio (H), the share of the population that is multidimensionally poor, and the intensity of poverty (A), the average share of weighted deprivations among the poor. MPI = H × A, so it is sensitive to both how many people are poor and how deeply.

The global MPI: 10 indicators, and India's current figure

The global MPI (UNDP-OPHI) uses 10 indicators: two under health (nutrition, child mortality), two under education (years of schooling, school attendance), and six under standard of living (cooking fuel, sanitation, drinking water, electricity, housing, assets), each weighted within its dimension. The 2025 global MPI report, "Overlapping Hardships: Poverty and Climate Hazards," released in October 2025 and covering 109 developing countries, gives India's most recent available figures, based on 2019/2021 survey data (the same NFHS-5 round used elsewhere in this chapter): a headcount of 16.4% multidimensionally poor, an intensity of deprivation of 42.0%, and an MPI value of 0.069, alongside a further 18.7% of the population classified as vulnerable to multidimensional poverty.

India's National MPI (NITI Aayog): the distinctly Indian exercise

Separate from the global UNDP-OPHI exercise, NITI Aayog developed India's own National MPI, with technical inputs from UNDP and OPHI, but using a methodology "Indianised" for domestic policy use: it expands the global MPI's 10 indicators to 12, adding maternal health (under the health dimension) and bank account access (under standard of living) to reflect priorities specific to Indian welfare programming. This national exercise has produced three successive readings, and mixing them up is a real risk:

• The Baseline Report (November 2021), using NFHS-4 (2015-16) data, found 24.85% of India multidimensionally poor.
• National Multidimensional Poverty Index: A Progress Review 2023 (released 17 July 2023), using NFHS-5 (2019-21) data, found poverty had fallen to 14.96%, meaning 13.5 crore people moved out of multidimensional poverty across those five years; the national MPI value itself nearly halved, from 0.117 to 0.066, and the intensity of poverty fell from 47% to 44%.
• A further Discussion Paper, "Multidimensional Poverty in India since 2005-06" (released 15 January 2024), extended the picture using projected estimates (not a fresh household survey) to claim a decline from 29.17% in 2013-14 to 11.28% in 2022-23, implying 24.82 crore people escaped multidimensional poverty over that nine-year span. Because this last figure rests on projection rather than a direct NFHS round, it should be cited as a modelled estimate, not treated as equivalent in evidentiary weight to the NFHS-anchored Progress Review figure of 14.96%.

The exam-relevant contrast to hold onto: the global MPI (16.4%, UNDP- OPHI, 10 indicators) and India's National MPI (14.96% for the same NFHS-5 survey round, NITI Aayog, 12 indicators) are two different measurements of the same country using different indicator baskets, not a single figure reported twice. Getting the sponsoring body, indicator count, or survey year swapped between the two is precisely the kind of trap this topic sets.

Human capital formation: investment in people as economic strategy

Human capital formation is the idea that spending on education, health, training and skill development is not consumption but investment, functionally parallel to investment in physical capital like machinery or infrastructure, because it raises a worker's future productivity and, in aggregate, the economy's output. The concept's modern foundation traces to economist Theodore W. Schultz, whose 1961 paper "Investment in Human Capital" (American Economic Review, Vol. 51) argued that growth in national output routinely outpaces the growth of land, labour hours and physical capital combined, and that investment in the skills and knowledge embedded in people is the major reason why. Gary Becker extended this into a fuller theoretical and empirical framework in his 1964 book Human Capital, formalising how individuals weigh education and training costs against the higher lifetime earnings such investment generates; his broader work on human capital contributed to his 1992 Nobel Memorial Prize in Economic Sciences. The practical Indian question this raises: since human capital is built through health and education spending exactly as physical capital is built through savings and investment, are India's health and education outlays sized and targeted as investment, or merely as welfare expenditure.

India's skilling architecture: NSDC, Skill India, PMKVY

India's institutional response to the human capital problem centres on a small number of interlocking bodies. The National Skill Development Corporation (NSDC), established in 2008 under the Ministry of Skill Development and Entrepreneurship (MSDE), is a public-private partnership whose stated mission is to enable a unified public-private skills ecosystem, catalysing private-sector-led training rather than running training itself.

The umbrella under which most central skilling sits is the Skill India Mission, and its flagship scheme is the Pradhan Mantri Kaushal Vikas Yojana (PMKVY), run by MSDE, not the Ministry of Labour and Employment, a distinction a real 2018 Prelims question tested directly. The current phase, PMKVY 4.0, delivers Short-Term Training (STT), Special Projects (SP), and Recognition of Prior Learning (RPL) for reskilling and upskilling, targets beneficiaries aged 15 to 59, and maps every certification to the National Skills Qualification Framework (NSQF).

The current institutional status, verified as of this writing: on 7 February 2025, the Union Cabinet approved a composite Central Sector Scheme, the "Skill India Programme" (SIP), running 2022-23 to 2025-26 with a combined outlay of ₹8,800 crore, folding PMKVY 4.0 (₹6,000 crore), the Pradhan Mantri National Apprenticeship Promotion Scheme (PM-NAPS) (₹1,942 crore, ages 14-35, subsidised apprentice stipends), and the Jan Shikshan Sansthan (JSS) scheme (₹858 crore, a low-cost, community-centred channel for women, rural youth and disadvantaged groups aged 15-45) into one scheme. Across these three components, MSDE reports over 2.27 crore beneficiaries to date. Treat this 2025 restructuring, not the older standalone PMKVY 4.0 framing, as the current baseline.

The demographic dividend: a window, not a guarantee

India's population age structure is unusually favourable, at least for now. Per the Economic Survey 2018-19's own analysis, India's working-age population has outnumbered its dependent population since around 2018, a bulge projected to persist until roughly 2055, with the working-age share peaking near 59% around 2041. This span is India's demographic dividend window: a period when a large share of the population is of working age relative to children and the elderly, which can translate into higher savings, higher labour supply and faster growth, provided that bulge of working-age people is actually employed productively.

That last clause is the entire analytical stake of the concept, and it is exactly what human capital formation and the skilling architecture above are meant to secure. A demographic dividend is not automatic: it converts into growth only if the working-age population is healthy, educated and skilled enough to be productively employed. Where that investment lags, the same population bulge becomes a demographic burden instead: a large cohort of underemployed or unemployable young people generating social and political strain rather than growth, on the same population, over the same timeline the dividend literature treats as an opportunity.

Quick revision points

• HDI dimensions and indicators: health (life expectancy at birth); education (expected years of schooling + mean years of schooling, averaged); standard of living (GNI per capita at PPP, not GDP).
• HDI formula: geometric mean of the three dimension indices since the 2010 Human Development Report (before that, an arithmetic mean); this switch removed perfect substitutability between dimensions.
• India's HDI (HDR 2025, 2023 data): rank 130 of 193, value 0.685, medium human development category (0.550 to 0.699).
• IHDI: discounts HDI for internal inequality; falls below HDI as inequality rises; inequality cuts 30.7% off India's HDI.
• GDI: ratio of female HDI to male HDI; a comparison of achievement levels using the same three HDI dimensions.
• GII: a genuinely separate index, not GDI's twin, covering reproductive health (maternal mortality ratio, adolescent birth rate), empowerment (parliamentary seats, secondary education) and labour market participation, by gender; higher GII means more disadvantage.
• Global MPI (UNDP-OPHI, since 2018, 10 indicators): India's headcount is 16.4% (2019/21 survey data, 2025 report). National MPI (NITI Aayog, 12 indicators, adds maternal health and bank accounts): 14.96% for the identical NFHS-5 (2019-21) round; baseline (NFHS-4, 2015-16) was 24.85%; a 2024 discussion paper projects (not surveys) 11.28% for 2022-23.
• Human capital formation: Schultz (1961) and Becker (1964) established investment in education, health and skills as functionally equivalent to investment in physical capital.
• PMKVY 4.0: MSDE's flagship (not Ministry of Labour and Employment), targets ages 15-59, delivers STT, Special Projects and RPL; since February 2025, folded with PM-NAPS and JSS into the composite Skill India Programme, ₹8,800 crore outlay, 2022-23 to 2025-26.
• Demographic dividend: India's working-age population has exceeded its dependent population since around 2018, peaking near 59% around 2041, with the window closing around 2055; it converts to growth only with matching human capital investment, and risks becoming a demographic burden without it.