Multidimensional Poverty Index (MPI): Alkire-Foster Method, NITI Aayog MPI, and India’s Progress

The Multidimensional Poverty Index is the most influential rethinking of poverty measurement in a generation. Where the dollar-a-day income line counts the poor by what they earn or consume, the Multidimensional Poverty Index counts the poor by the simultaneous deprivations they suffer — bad health, missing schooling, no toilet, no clean cooking fuel, no electricity. Developed jointly by the Oxford Poverty and Human Development Initiative (OPHI) and the United Nations Development Programme (UNDP) and first published in the 2010 Human Development Report, the MPI now anchors poverty discourse in more than 100 countries, including India, which since 2021 produces its own National MPI through NITI Aayog. For UPSC GS-II and GS-III aspirants, the Multidimensional Poverty Index is the gateway concept for poverty, inclusive growth, and the social-sector debate, and it sits alongside the Human Development Index as a core human development measure.

Why a Multidimensional Index at All

The income-based poverty line, in India most famously associated with the Tendulkar and Rangarajan committees, asks a single question — does this household consume more or less than a defined rupee value? It does not ask whether a household has a working toilet, whether its children are stunted, or whether a girl child completed eight years of schooling. Yet decades of survey data show that income poverty and these deprivations move together imperfectly. A household above the income line can still be deprived on multiple human-development fronts, and a household marginally below it can have a healthy, educated child.

The capability approach pioneered by Amartya Sen reframed poverty as the deprivation of basic capabilities. The MPI operationalises that approach by counting overlapping deprivations directly, rather than treating income as a proxy.

The OPHI-UNDP Global MPI

The global Multidimensional Poverty Index is built on three dimensions and ten indicators, drawn directly from data routinely collected in the Demographic and Health Survey (DHS) and the Multiple Indicator Cluster Survey (MICS) — and in India’s case, the National Family Health Survey (NFHS).

Three Dimensions, Ten Indicators

Each dimension is weighted equally at one-third, and within each dimension the indicators are equally weighted. A person is identified as multidimensionally poor if they are deprived in at least one-third (33.33 percent) of the weighted indicators.

The Alkire-Foster Method

The MPI is built on the Alkire-Foster method, named after the two economists Sabina Alkire (OPHI’s director) and James Foster, who developed it in 2007–10. The method has two distinctive features.

Dual Cutoff Identification

The first cutoff is at the indicator level — what counts as deprivation in nutrition, schooling, sanitation, and so on. The second cutoff is at the person level — what fraction of weighted deprivations must a person suffer to be called multidimensionally poor. The global MPI sets this poverty cutoff (k) at one-third. India’s National MPI uses the same threshold.

M0 — The Adjusted Headcount Ratio

The MPI value is the product of two numbers:

MPI = H × A

Where:

• H (Headcount Ratio) is the share of people who are multidimensionally poor.
• A (Intensity) is the average share of weighted deprivations that the poor experience.

This M0 measure — also called the Adjusted Headcount Ratio — has a powerful property: it falls not only when people exit poverty (H drops) but also when the poor become less poor (A drops). That dual sensitivity is what makes the Alkire-Foster method genuinely informative for policy. It is also fully decomposable by subgroup, indicator, and geography, allowing policymakers to identify exactly which deprivation is dragging which region.

India’s National MPI by NITI Aayog

India’s National Multidimensional Poverty Index, published by NITI Aayog, adapts the global MPI to Indian conditions. The first baseline report appeared in 2021 using NFHS-4 (2015–16) data, and the second progress report in July 2023 used NFHS-5 (2019–21).

India’s MPI Structure — 12 Indicators

NITI Aayog retains the three dimensions but uses twelve indicators rather than ten, adding two indicators uniquely relevant to India:

• Maternal health (under the health dimension)
• Bank account (under living standards)

The other ten indicators map to the global MPI, with some thresholds tightened for Indian standards.

Headline Findings — 2023 Progress Report

The 2023 National MPI Progress Report announced a dramatic decline:

• MPI headcount ratio fell from 24.85 percent in 2015–16 to 14.96 percent in 2019–21.
• Intensity declined from 47.14 percent to 44.39 percent.
• MPI value (M0) dropped from 0.117 to 0.066.

That headline 14.96 percent figure means roughly one in seven Indians remained multidimensionally poor as of NFHS-5. A subsequent NITI Aayog discussion paper estimated that around 24.82 crore (248 million) Indians exited multidimensional poverty between 2015–16 and 2022–23, with the rural poor accounting for the bulk of the decline.

Combined with the earlier 2005-06 to 2015-16 decline calculated by OPHI, the cumulative claim is that around 415 million Indians moved out of multidimensional poverty between 2005-06 and 2019-21 — the largest such reduction in any country over that period.

State Rankings

Bihar, Jharkhand, Uttar Pradesh, Madhya Pradesh, and Meghalaya remain the states with the highest MPI headcount ratios, even though all of them have shown faster-than-average improvement. Kerala, Tamil Nadu, Punjab, Goa, and Sikkim consistently lead the rankings with multidimensional poverty under 1 percent.

Global MPI — India’s Position

In the OPHI-UNDP global MPI 2025, India’s MPI headcount ratio (using slightly different international cutoffs) is around 16.4 percent and the MPI value is roughly 0.069. Those figures are unchanged from the previous edition for a reason worth stating plainly: India’s global MPI still rests on NFHS-5 fieldwork from 2019–21, so the number reflects the survey vintage, not a stalled decade. About 234 million Indians are estimated to be multidimensionally poor by the global definition, the largest such absolute number in any single country, even though India’s MPI headcount ratio is now well below several other low-income countries.

Drivers of the Decline

Several flagship Indian schemes mapped neatly onto specific MPI indicators and almost certainly contributed to the decline:

• Swachh Bharat Mission — sanitation indicator
• Pradhan Mantri Awas Yojana — housing indicator
• Saubhagya — electricity indicator
• Pradhan Mantri Ujjwala Yojana — cooking fuel indicator
• Jal Jeevan Mission — drinking water indicator
• Pradhan Mantri Jan Dhan Yojana — bank account indicator
• POSHAN Abhiyaan — nutrition indicator
• Ayushman Bharat — health access (indirect)

The MPI critics rightly point out that several of these are infrastructure or asset indicators that, once built, do not flip back — sanitation coverage cannot fall once toilets are built. But that is also the point: multidimensional poverty captures lasting structural changes that income volatility does not.

Criticisms of the MPI

The Multidimensional Poverty Index has not escaped methodological criticism:

• Equal weighting of dimensions and indicators is a normative choice without strong empirical basis.
• Cutoff arbitrariness — the 33.33 percent threshold has no theoretical justification.
• Cardinal interpretation — MPI values are sometimes treated as ordinal rankings but used cardinally in policy.
• Data lag — NFHS rounds are five years apart, so the MPI cannot capture quick shocks like Covid-19.
• Indicator substitution — an indicator like “assets” can be passed by owning a fridge but failing on nutrition.

Despite these critiques, the MPI is treated as a major advance over income-only poverty lines, including by the World Bank, which since 2018 publishes a Multidimensional Poverty Measure alongside its $2.15-a-day extreme poverty line.

MPI, HDI and SDG 1

The MPI is the operational counterpart of SDG 1.2 — “reduce at least by half the proportion of men, women, and children of all ages living in poverty in all its dimensions.” It also relates closely to the Gini Coefficient, although the MPI measures deprivation while the Gini measures distributional inequality. The MPI is the household-level cousin of the Human Development Index, and the global MPI report is now released as part of UNDP’s Human Development Report Office calendar.

The Way Forward

India’s next NFHS round will be the first stress test of whether the post-2021 decline in multidimensional poverty has been sustained through the pandemic recovery and the slow normalisation of rural wages. The bigger task is filling the residual pockets — the 14.96 percent translates to roughly 200 million Indians who still face overlapping deprivations every day. Targeted convergence of welfare schemes around the high-MPI districts, the strategy NITI Aayog has used in its Aspirational Districts Programme, remains the single most evidence-aligned approach to reaching zero multidimensional poverty by 2030.

Frequently Asked Questions

PDF