OFFICIAL STATISTICS SPECIALLY DESIGNED FOR INDIAN STATISTICAL SERVICE (ISS) AND OTHER GOVERNMENT & PRIVATE SECTOR STATISTICAL EXAMINATIONS
Get the complete book, written for ISS aspirants and every statistical exam that tests Indian official statistics.
Buy the book on Amazon.inNational Multidimensional Poverty Index (MPI) of India: Alkire-Foster Method and Exam Questions for ISS
Headlines about crores of people moving out of poverty often cite the National Multidimensional Poverty Index (MPI), published by NITI Aayog. Unlike an income-based poverty line, it looks at deprivations in health, education and living standards together. It is a rising favourite of Indian Statistical Service (ISS) paper setters and also appears in RBI, NABARD, SSC and university statistics exams. This guide gives you the method, the weights and three exam-style questions.
1. What is the Multidimensional Poverty Index?
The MPI measures poverty as the share of people who are deprived in several non-monetary indicators at the same time, and how intense those deprivations are. India's national MPI follows the Alkire-Foster methodology developed by OPHI and is published by NITI Aayog.
Why examiners love this topic
The MPI links index construction, weighting, cutoffs and decomposition in one place. Numerical questions are easy to set and easy to get wrong.
2. Who publishes it and what data does it use?
NITI Aayog releases the national MPI. The data come mainly from the National Family Health Survey (NFHS), so reports are tied to survey rounds such as NFHS-4 (2015-16) and NFHS-5 (2019-21). NITI Aayog's 2023 review reported the headcount ratio falling from about 24.85% to about 14.96% between these two rounds. Check the latest release before quoting figures.
Dimensions
Three, each with equal weight of 1/3: Health, Education and Standard of Living.
Indicators
Twelve in all: 3 in health, 2 in education and 7 in standard of living.
Poverty cutoff
A person is multidimensionally poor if the weighted deprivation score is 1/3 (33.33%) or more.
Exam tip
The 1/3 cutoff applies to the weighted deprivation score, not to the number of indicators. A person deprived only in school attendance (weight 1/6) is not poor on that alone.
3. The formula behind the MPI
The Alkire-Foster method has two steps: identify who is poor, then aggregate. The MPI combines incidence and intensity.
Here n is the population, q is the number of poor people, and ci is the weighted deprivation score of the i-th poor person. H is the headcount ratio (incidence) and A is the average intensity of deprivation among the poor.
- Strength: it can be decomposed by region, group or indicator to show where poverty comes from.
- Weakness: weights and cutoffs are normative choices, and most indicators are simply deprived or not deprived.
- Property to remember: it is a non-monetary measure. It does not use income or consumption.
4. Indicator weights at a glance
Weights are equal across dimensions but unequal across indicators. Each education indicator carries the highest individual weight, and each standard of living indicator the lowest.
Weights in per cent (total = 100): Health 3 × 11.11, Education 2 × 16.67, Standard of living 7 × 4.76. The seven standard of living indicators are cooking fuel, sanitation, drinking water, electricity, housing, assets and bank account.
5. Global MPI vs India's national MPI: the comparison that keeps coming up
| Basis | Global MPI (UNDP-OPHI) | India's national MPI |
|---|---|---|
| Dimensions | 3: health, education, living standards | 3: the same |
| Indicators | 10 | 12 |
| Health | Nutrition, child mortality | Nutrition, child and adolescent mortality, maternal health |
| Standard of living | 6 indicators | 7 indicators (adds bank account) |
| Poverty cutoff | 1/3 | 1/3 |
| Main data source | DHS, MICS or national surveys | NFHS |
6. Common mistakes students make
- Treating the MPI as the headcount ratio alone. MPI = H × A.
- Saying a person is poor if deprived in any one indicator. The rule is a weighted score of 1/3 or more.
- Assuming all 12 indicators have equal weight. Equal weights apply to dimensions, not indicators.
- Calling it an income poverty measure. The MPI is non-monetary.
- Attributing the index to the Planning Commission or NSO. NITI Aayog publishes it.
7. Three important questions on the MPI
Try each one on your own first, then tap the answer.
Name the three dimensions of the national MPI and state the cutoff used to identify the multidimensionally poor.
Show answer
The dimensions are Health, Education and Standard of Living, each with weight 1/3. A person is identified as multidimensionally poor if the weighted sum of deprivations is 1/3 (33.33%) or more.
A person is deprived in nutrition, years of schooling, cooking fuel, sanitation, housing and bank account. Using the national MPI weights, find the deprivation score and say whether the person is poor. In a sample of 200 people, 50 are poor with an average deprivation score of 45%. Compute H, A and the MPI.
Show answer
Score = 1/9 (nutrition) + 1/6 (schooling) + 4 × 1/21 (fuel, sanitation, housing, bank account) = 0.1111 + 0.1667 + 0.1905 = 0.4683.
0.4683 ≥ 0.3333, so the person is multidimensionally poor.
H = 50 / 200 = 25%. A = 45%. MPI = 0.25 × 0.45 = 0.1125.
Explain the Alkire-Foster method used in India's national MPI. How does it differ from income-based poverty measurement, and what are its limitations?
Show answer
Write your answer in four parts. (a) Identification: set a deprivation cutoff for each of the 12 indicators, weight the indicators (1/3 per dimension), compute each person's deprivation score and apply the poverty cutoff of 1/3. (b) Aggregation: compute the headcount ratio H and intensity A; MPI = H × A. It can be decomposed by region or indicator. (c) Difference from income poverty: income-based measures compare consumption with a poverty line, whereas the MPI captures several non-monetary deprivations that occur together and is not affected by price changes. (d) Limitations: weights and cutoffs are normative, indicators are mostly binary, results are sensitive to these choices, and NFHS data arrive only every few years. Conclude that the MPI complements, and does not replace, monetary poverty estimates.
Key takeaways
- The national MPI has 3 dimensions and 12 indicators, and is published by NITI Aayog using NFHS data.
- A person is poor if the weighted deprivation score is 1/3 or more.
- MPI = H × A: incidence multiplied by intensity.
- Equal weights are across dimensions (1/3 each), not across the 12 indicators.
Everything on Official Statistics, in one book
The MPI is just one chapter of the syllabus. If you want the complete picture, this book is built for you.
- Indian official statistical system: MoSPI, NSO, NSC and the data they produce
- National accounts, GDP, IIP, price indices and labour statistics
- NSS surveys, sampling designs and data quality concepts
- Written specially for the ISS exam, and useful for other government and private sector statistical exams
- Concept-focused notes to revise quickly before the exam
Official Statistics for ISS and other Government & Private Sector Statistical Examinations