Multidimensional Poverty Index: How NFHS Data is Shaping India’s Fight Against Poverty

What is MPI?

Poverty is a multidimensional challenge that extends beyond income-based measures. The Multidimensional Poverty Index (MPI), developed by Sabina Alkire and James Foster in 2010, was adopted by UNDP in its Human Development Report to capture overlapping deprivations in health, education, and living standards (Source). This index provides a nuanced understanding of poverty and helps policymakers design targeted interventions.

In India, NITI Aayog, in collaboration with UNDP and Oxford Poverty and Human Development Initiative (OPHI), developed the National MPI to provide a comprehensive measure of poverty at national, state, and district levels. It aligns with SDG Target 1.2, which aims to reduce poverty in all its dimensions by half by 2030.

The National MPI: Progress Review 2023 (Download the report here) presents estimates for 36 States & UTs and 707 districts based on NFHS-5 (2019-21) data, comparing progress with NFHS-4 (2015-16).

But Wait! What is NFHS?

The National Family Health Survey (NFHS) is a large-scale, multi-round survey conducted in India to collect essential data on population, health, and nutrition. Initiated in 1992-93, the first round, NFHS-1, aimed to provide national and state-level estimates on fertility, family planning, and maternal and child health indicators. The survey was conducted under the stewardship of the Ministry of Health and Family Welfare (MoHFW), with the International Institute for Population Sciences (IIPS) in Mumbai designated as the nodal agency. Technical assistance was provided by international organizations, including the United States Agency for International Development (USAID). Over the years, subsequent rounds—NFHS-2 (1998-99), NFHS-3 (2005-06), NFHS-4 (2015-16), and NFHS-5 (2019-21)—have expanded in scope and coverage, incorporating new topics such as reproductive health, HIV prevalence, and domestic violence, while also enhancing data quality and reliability through methodological improvements.

MPI Measurement

MPI measures poverty across multiple dimensions—Health, Education, and Standard of Living—instead of just income. Each household is assessed based on 12 indicators, each with a specific weight as mentioned in the following table:

To understand how the Multidimensional Poverty Index (MPI) is calculated, let's take an example of three households (HH1, HH2, HH3), each with multiple family members.

The following table shows whether each household is deprived (1) or not deprived (0) in each indicator, along with the assigned weights. The deprivation score for each household is calculated by summing up the weighted deprivations as shown in the following table:

The table presents the deprivation scores for three households, illustrating how multidimensional poverty is assessed. Each household is evaluated against 12 indicators, with deprivations assigned weights corresponding to their importance in measuring poverty. A household's total deprivation score is the sum of its weighted deprivations, reflecting the extent to which it experiences multidimensional poverty. In this example, HH1 has a deprivation score of 72.8%, HH2 has 52.6%, and HH3 has 21.4%.

A household is considered multidimensionally poor if its deprivation score exceeds 33%. Based on this threshold, HH1 and HH2 are classified as poor, while HH3 is not. Since household size varies, the next step is to calculate the headcount ratio, which represents the proportion of individuals who are multidimensionally poor in the total population. HH1 consists of seven members, HH2 has five, and HH3 has four, making the total population 16. With HH1 and HH2 being poor, the number of individuals in poverty is 12. The headcount ratio (H) is thus calculated as:

H = 12 / 16 i.e. H = 0.75

This means that 75% of the individuals in this dataset are multidimensionally poor. However, poverty is not just about how many people are affected but also the depth of their deprivation. To account for this, the intensity of poverty (A)measures the average proportion of deprivations experienced by poor individuals. It is calculated by summing the deprivation scores of poor households and dividing by the total number of poor individuals:

A=(7×0.728)+(5×0.526)12=0.64A=12(7×0.728)+(5×0.526)=0.64

On average, a poor individual is deprived in 64% of the weighted indicators. The final MPI score is derived by multiplying the headcount ratio (H) with the intensity of poverty (A):

MPI=H×A=0.75×0.64=0.48MPI=H×A=0.75×0.64=0.48

This MPI score of 0.48 indicates significant multidimensional poverty in the given dataset. The advantage of MPI over income-based measures is its ability to highlight specific areas where deprivation is concentrated, enabling targeted policy interventions. By tracking both the incidence and severity of poverty, MPI offers a more nuanced understanding of economic and social inequality, informing strategies for poverty alleviation at multiple levels.

Okay, So, Details of NFHS-5?

NFHS-5, which was released on November 24, 2021, collected data between 2019 and 2021. This means the data is not entirely from 2021 but spans multiple years, with some indicators reflecting responses from as early as mid-2019. This is critical to understand because any conclusions drawn should consider the time lag, especially for rapidly changing indicators such as health interventions, economic conditions, or program effectiveness. So, it is not a real-time dataset, and the cut-off is not strictly 2021, making it essential to interpret trends with caution, particularly in post-pandemic analyses.

NFHS 5: Approach

The National Family Health Survey (NFHS-5) (2019-21) used a stratified two-stage sampling approach, covering 707 districts across 28 states and 8 union territories. Rural and urban areas were separately stratified, with villages and Census Enumeration Blocks (CEBs) as Primary Sampling Units (PSUs), selected using probability proportional to size (PPS). Within each PSU, 22 households were systematically chosen. The survey included four questionnaires—Household, Woman, Man, and Biomarker—administered via Computer-Assisted Personal Interviewing (CAPI) in 18 regional languages. Data collection was led by 1,061 field teams, each comprising interviewers, supervisors, and health investigators, with biomarker tests measuring anemia, blood glucose, and hypertension. SyncCloud technology enabled real-time data monitoring, with spot checks, back-checks, and automated validation ensuring accuracy. Field staff underwent centralized and state-level training, following a Training of Trainers (ToT) model. Conducted by the International Institute for Population Sciences (IIPS), Mumbai, with technical support from ICF (USA) under the Demographic and Health Surveys (DHS) program, NFHS-5 provides district and state-level estimates, maintaining global survey standards.

NFHS 5: The fine lines

The initial phase of NFHS-5 findings was released on December 12, 2020[1], covering 22 States and Union Territories. The comprehensive national report was subsequently released on May 5, 2022[2]. This means the data is not entirely from 2021 but spans multiple years, with some indicators reflecting responses from as early as mid-2019. This is critical to understand because any conclusions drawn should consider the time lag, especially for rapidly changing indicators such as health interventions, economic conditions, or program effectiveness. So, it is not a real-time dataset, and the cut-off is not strictly 2021, making it essential to interpret trends with caution, particularly in post-pandemic analyses.

Great, So India's Standing in MPI?

India is reducing multidimensional poverty, with the most significant improvements observed in electricity access (73.1%), sanitation (41.9%), and cooking fuel (24.9%). The indicator wise improvement is mentioned in the following table. However, progress in nutrition and school attendance has been much slower, with a substantial proportion of the population still facing deprivation in these areas. Bank account ownership has remained stagnant at 3.56%, indicating limited progress in financial inclusion (Or, maybe it relies on an indicator for which the periodicity of measurement is beyond the periodicity of two consecutive NFHS).

Among the areas where India has performed well, electricity access stands out with a 73.1% improvement, bringing deprivation levels down to just 3.27%. Similarly, sanitation has seen a 42% reduction in deprivation, reflecting the success of programs like the Swachh Bharat Abhiyan(?). The availability of clean cooking fuel has also improved significantly, with a 25% reduction in deprivation, pointing to the effectiveness(?) of efforts to promote LPG usage across rural and urban areas.

Despite these advancements, critical gaps remain. Nutrition continues to be the largest contributor to MPI (29.86%), with 31.52% of people still experiencing deprivation. The education sector, particularly years of schooling and school attendance, accounts for over 16% of the MPI score, signaling that progress in literacy and education accessibility has been slower than needed. Additionally, asset ownership and financial inclusion have seen limited improvements, with persistent gaps in economic security.

Overall, India has made significant progress in reducing multidimensional poverty, particularly in improving infrastructure-related indicators such as electricity, sanitation, and fuel access. However, nutrition, education, and financial security remain critical challenges, requiring sustained efforts and targeted interventions to ensure holistic poverty reduction.

[1] Press Information Bureau. (2020, December 15). National Family Health Survey-5. Government of India. https://pib.gov.in/Pressreleaseshare.aspx?PRID=1680702

[2] Press Information Bureau. (2022, May 5). Union Health Minister Dr Mansukh Mandaviya releases NFHS-5 report. Government of India. https://pib.gov.in/PressReleaseIframePage.aspx?PRID=1823047