The 1/3 problem in India’s poverty estimation

The Multidimensional Poverty Index (MPI) marked a significant departure from traditional approaches to measuring poverty.

Instead of focusing narrowly on income, the MPI captures a wider spectrum of deprivations that individuals and households experience across health, education, and living standards. In India, the national MPI, developed by NITI Aayog in partnership with the United Nations Development Programme (UNDP) and the Oxford Poverty and Human Development Initiative (OPHI), has emerged as a significant method for monitoring progress in poverty reduction.

Based on the household-level data from the National Family Health Survey (NFHS), the index covers 12 indicators ranging from education to health to standard of living, each equally weighted within its dimension, to construct a composite picture of deprivation. A household is identified as multidimensionally poor if it suffers deprivations in at least one-third of the total weighted indicators. This threshold, known as k = 1/3, is rooted in the Alkire-Foster methodology and ensures that poverty measurement captures both the breadth and intensity of deprivation.

Yet, this seemingly simple cutoff has increasingly attracted criticism for its arbitrariness and lack of empirical grounding, raising serious questions about the validity of India’s MPI estimates. The choice of one-third deprivation (k = 1/3) rests more on normative judgement than on rigorous empirical justification. As a result, even minor adjustments to this threshold can produce large shifts in measured poverty, which alter both headcount ratios and narratives of current estimates. For example, the reported fall in multidimensional poverty from 29.17% in 2013-14 to 11.28% in 2022-23 may, at least in part, reflect the sensitivity of results to methodological choices rather than an unambiguous structural transformation. This subjectivity creates space for bias, as the choice of technical parameters and their equal weights may be shaped intentionally or unintentionally by political goals to present rapid progress, rather than by an objective or the current picture of deprivation on the ground.

Let’s understand the problem this causes on the ground. Even going by the 1/3 rule – which states that anybody with a score above 33.33% is poor – its application in India makes a gruesome mistake due to its sharp cutoff. In a country of billions, this is worrisome. Consider somebody scoring 32% and another scoring 34%. While computationally there is a gap of 2%, it is unlikely that any significant difference would be observed on the ground between the two households. Therefore, the cut-off, because of its rigidity, often does an injustice to the households just below the threshold. In India, this could mean that tens of millions of people who are one step away from poverty are not seen as a priority.

The fact that governments expect back-patting for poverty reduction based on these numbers is absolutely contrary to what one would expect from the MPI. Remember that the MPI is largely established based on the Capability Approach, developed by Nobel Prize-winning economist Prof. Amartya Sen. This approach is different also because it focuses on the freedom and capability of individuals to aspire to and achieve a valued life. However, in a welfare state where the government provides many necessities like toilets, roads, houses, and rations, and these provisions push households below the 1/3 cutoff, two aspects are ignored that lead to an underestimation of poverty: (i) These households were initially incapable of fulfilling those necessities themselves. By counting the government-provided necessities, we risk underestimating their underlying economic vulnerability. (ii) By sticking rigidly to the 1/3 rule, this approach exacerbates the peril of underestimation. This is why we should categorise, actually the poor and officially the poor.

A flawed foundation

Actual poverty, in this context, goes beyond mere access to Standard of Living (SOL) indicators such as sanitation facilities; it emphasises the actual utilisation and sustained use of these resources to ensure meaningful improvements in well-being. A household may have a toilet constructed through government schemes, but if cultural and caste barriers, maintenance issues, or lack of water prevent its regular use, the deprivation persists in practice, rendering the official classification misleading.

The obvious on/off switch created by the 1/3 cutoff results in two groups of people: the poor and the not-poor. Most importantly, an individual designated as poor with 34% of indicators and someone with 94% are categorised identically – simply as “poor.” Consequently, the depth of multidimensional poverty is lost. This is not to blame the MPI itself, but this cutoff thinking impacts policymaking significantly. The recent governmental focus on improving national and international rankings shifts the policy target. The fact that it is easier to bring someone at 34% below the 33% line by simply providing a toilet makes it more efficient for the government to reduce the poverty headcount. However, this approach shifts the focus away from the individual at 94% deprivation, who requires more immediate and comprehensive attention.

Recent research by Bonnerjee (2025) has pointed to the inherent instability of the k =1/3 cutoff. Data-driven approaches suggest more stable cut-offs between 0.22 and 0.24. Applying these thresholds yields MPI figures that are 69-127% higher than the official estimates. Sensitivity checks using alternative cutoffs and weighting schemes reinforce the point that methodological choices can significantly distort poverty measurement, especially in a country as diverse as India, where deprivation patterns vary widely across regions, castes, and household types (Alkire et al.,2019).

Questioning the 1/3 cutoff is not merely an academic exercise; it has profound implications for policy formulation, resource allocation, and India’s commitment to the Sustainable Development Goals. If the MPI’s foundational assumptions are flawed, the celebrated narrative of lifting millions out of poverty may mask persistent inequalities. More importantly, such mismeasurement of countries’ important variables diverts attention from the structural challenges that hinder inclusive development. The growth and welfare must reach marginalised communities and households that remain trapped in multidimensional deprivation.

(Kaibalyapati is a senior research fellow at CESP, ISEC, Bengaluru; Jadhav is an assistant professor at the Centre for Economic and Social Studies [CESS], Hyderabad)