Multidimensional Poverty Index (MPI)

Poverty is not only about money. A family can earn above the poverty line and still have no clean cooking fuel, no toilet or no school nearby. The Multidimensional Poverty Index (MPI) counts those missing basics. It counts them for each person at the same time.

1.1 What does the Global MPI measure?

The Global MPI measures acute multidimensional poverty across more than 100 developing countries. Acute means severe. It looks at each person’s overlapping deprivations across 10 indicators in 3 equally weighted dimensions.

• Health: 2 indicators, nutrition and child mortality.
• Education: 2 indicators, years of schooling and school attendance.
• Standard of living: 6 indicators, such as cooking fuel, sanitation and electricity.

All the indicators for one country come from the same household survey. The MPI runs from 0 to 1. A higher value means more multidimensional poverty.

1.2 Who built it, and who runs it now?

The 2010 Human Development Report introduced the MPI. Since 2018 UNDP’s Human Development Report Office (HDRO) and the Oxford Poverty and Human Development Initiative (OPHI) publish it together. The original method is described in a 2010 paper by Alkire and Santos. Sabina Alkire is the director of OPHI.

1.3 How is “poor” decided in one line?

Each indicator has a weight. A person adds up the weights of every indicator where they are deprived. This is the deprivation score. A score of 1/3 or more means the person is multidimensionally poor. A score from 1/5 up to less than 1/3 means vulnerable. A score of 1/2 or more means severe poverty.

Both the Global MPI and India’s National MPI use the Alkire-Foster (AF) method. NITI Aayog calls it a “dual-cutoff counting method”. There are two cutoffs, and each does a different job.

2.1 What are the two cutoffs?

The first cutoff works inside one indicator. It decides whether a household is deprived in that indicator. For example, a household that cooks with wood is deprived in cooking fuel.

The second cutoff works across all indicators. It asks how much of the total weight a person is deprived in. India’s National MPI uses the same cutoff as the global MPI, 33.33 percent. Only past that line does the person count as poor.

2.2 How does the counting work? A made-up example

Take a household that is deprived in four indicators. Use India’s weights. Nutrition counts 1/6. Cooking fuel, sanitation and housing count 1/21 each.

• Score: 1/6 + 3/21 = about 0.31.
• That is below 1/3 (about 0.33). The household is not counted as poor.

Now add one more deprivation, electricity, which counts 1/21.

• Score: 1/6 + 4/21 = about 0.36.
• That is above 1/3. The household is counted as poor.

This example is invented to show the sums. The weights come from NITI Aayog.

2.3 What are H, A and the MPI value?

The MPI value is built from two parts.

• H (headcount ratio): the share of people who are poor. It answers “how many are poor?”.
• A (intensity): the average share of weighted deprivations among the poor.
• MPI = H x A.

India’s 2019-21 figures show the sum. H was 14.96 percent and A was 44.39 percent. The MPI value was 0.066.

2.4 Why is this different from income poverty?

Income poverty asks one question: is a person’s spending above or below a money line? The MPI asks which basic things are missing. UNDP gives India as an example. Its 2019/2021 MPI headcount is 11.1 percentage points higher than the share below the $3.00 a day line. Some people above the money line still lack health, schooling or living standards.

India did not just copy the Global MPI. NITI Aayog built a national version on the same method. It keeps the global list and adds two indicators.

3.1 What does it keep, and what does it add?

The National MPI retains the ten indicators of the global model. It adds Maternal Health (in the Health dimension) and Bank Account (in the Standard of Living dimension). That makes 12 indicators in the same three equally weighted dimensions.

NITI Aayog computes it. It uses data from the National Family Health Survey (NFHS).

3.2 Why were these two added?

The baseline report says a bank-account indicator gives the national MPI more links with the Sustainable Development Goals. It also notes that Rwanda and Sierra Leone include a bank-account indicator in their national MPIs.

In India’s list, Child and Adolescent Mortality and Maternal Health carry 1/12 each. Together they carry the 1/6 that child mortality has in the global list. NITI Aayog says this shared weight lets Nutrition keep its original share.

3.3 How do the weights compare?

3.4 What is the common trap?

Maternal Health and Bank Account exist only in India’s National MPI. They are not part of the Global MPI’s 10 indicators. Exam questions often claim that they are shared. They are not.

All 12 indicators sit inside three dimensions, each worth one third. A household is called deprived in an indicator when it meets the rule below.

4.1 What are the rules, indicator by indicator?

4.2 Why does one person’s nutrition count for the whole household?

On the Nutrition indicator, one undernourished member is enough to mark the whole household deprived. NITI Aayog explains the reason. The ill effects of undernourishment on one member reach other members of the same household.

Adults are undernourished if their Body Mass Index (BMI) is below 18.5 kg/m2. This BMI rule belongs to Nutrition, not to Maternal Health.

4.3 What is Maternal Health, then?

Maternal Health is about care around childbirth. It looks at women who gave birth in the last five years. A household is deprived if one of them had fewer than 4 antenatal care visits for the latest birth. It is also deprived if she had no trained skilled help at that birth. An antenatal visit is a health check during pregnancy.

4.4 Which indicator contributed most to India’s MPI value in 2015-16?

In the baseline report, Nutrition contributed 28.14 percent of India’s MPI value. NITI Aayog says it contributes nearly one-third of multidimensional poverty and is arguably one of the most important indicators.

Three NITI Aayog documents track the fall. They use different survey years. Learn each one with its own dates.

5.1 What did the baseline report (November 2021) find?

The baseline report was India’s first national MPI. It used NFHS-4 data from 2015-16, not NFHS-5. PIB says it was launched in November 2021.

• Headcount: 25.01 percent of people were multidimensionally poor.
• Intensity: 47.13 percent. MPI value: 0.118.
• Rural: 32.75 percent. Urban: 8.81 percent.

5.2 What did the 2023 progress review find?

The July 2023 review used NFHS-5 (2019-21) and compared it with NFHS-4 (2015-16). Vice-Chairman Suman Bery released it.

• Headcount fell from 24.85 to 14.96 percent, a drop of 9.89 percentage points.
• About 13.5 crore people moved out of poverty.
• Rural poverty fell from 32.59 to 19.28 percent. Urban poverty fell from 8.65 to 5.27 percent.
• The MPI value nearly halved, from 0.117 to 0.066.
• Uttar Pradesh had the largest fall in the number of poor, at 3.43 crore.

Look at 2015-16 again. The 2021 report says 25.01 percent. The 2023 review says 24.85 percent. The reports do not explain the gap in the text read for this lesson. In an exam, use the figure of the report the question names.

5.3 What did the 2024 discussion paper find?

NITI Aayog released the discussion paper “Multidimensional Poverty in India since 2005-06” on 15 January 2024. Prof Ramesh Chand released it. OPHI and UNDP gave technical inputs.

• Poverty fell from 29.17 percent in 2013-14 to 11.28 percent in 2022-23. That is a drop of 17.89 percentage points.
• About 24.82 crore people escaped multidimensional poverty in nine years.
• Uttar Pradesh led with 5.94 crore, then Bihar (3.77 crore), Madhya Pradesh (2.30 crore) and Rajasthan (1.87 crore).

5.4 Are the 2013-14 and 2022-23 figures surveyed?

No. Both are projected estimates. PIB says projected estimates were used because of data limits for these years. The paper computed 2013-14 from the 2005-06 estimate, using a yearly fall of 7.69 percent. It projected 2022-23 by applying the 2015-16 to 2019-21 rate of fall.

The surveyed points are 2005-06 (55.34 percent), 2015-16 (24.85 percent) and 2019-21 (14.96 percent). The paper also says the fall was faster after 2015-16. The yearly rate of decline was 10.66 percent from 2015-16 to 2019-21 and 7.69 percent from 2005-06 to 2015-16.

5.5 How do the three “people who escaped poverty” numbers differ?

Do not mix up these counts. Each covers a different period and a different index.

• 13.5 crore: National MPI, 2015-16 to 2019-21 (NITI Aayog, 2023).
• 24.82 crore: National MPI, 2013-14 to 2022-23, with projected end years (NITI Aayog, 2024).
• 415 million (41.5 crore): India’s fall in the Global MPI, 2005/06 to 2019/21 (UNDP and OPHI, 2023 report). The incidence fell from 55.1 to 16.4 percent.

UNDP and OPHI released the 2025 Global MPI on 17 October 2025. Its title is “Overlapping Hardships: Poverty and Climate Hazards”.

6.1 What is new in the report?

For the first time, the report overlays data on climate hazards and multidimensional poverty. It checks how exposed poor people are to four hazards: high heat, drought, floods and air pollution.

• 109 countries are covered, with 6.3 billion people. This update adds data from 13 countries.
• 1.1 billion of them are poor, about 18.3 percent.
• 887 million poor people live in regions with at least one of the four hazards.
• 309 million face three or four hazards at once.

6.2 What does it say about India?

The latest survey data for India refer to 2019/2021. UNDP reports these results.

• Headcount: 16.4 percent of people are multidimensionally poor.
• Intensity: 42.0 percent. MPI value: 0.069.
• Vulnerable: 18.7 percent. Severe poverty: 4.2 percent.
• Rural MPI: 0.089. Urban MPI: 0.023 (OPHI briefing).

For comparison, UNDP gives 0.067 for Bangladesh and 0.198 for Pakistan.

6.3 Did India get a rank?

Some notes say India ranked 66th of 109 countries. UNDP does not publish such a rank. Its FAQ says the survey dates differ across countries, which limits direct comparison. It adds, “This is the reason why we do not rank countries based on the MPI value.” Quote the MPI value 0.069, not a rank.

6.4 Why do India’s national and global numbers differ?

NITI Aayog’s 2019-21 headcount is 14.96 percent (12 indicators). UNDP’s 2019/2021 headcount is 16.4 percent (10 indicators). The two indices use different indicator lists and methods. Compare numbers only within the same index.

Two kinds of news matter. One is new survey data, NFHS-6. The other is how often exams ask about the MPI.

7.1 What does NFHS-6 say about MPI inputs?

The Health Ministry released NFHS-6 on 29 May 2026. The survey ran in 2023-24. The fact sheet compares it with NFHS-5 (2019-21). Several results touch MPI indicators.

These are survey results. They are not a new MPI. The sources read for this lesson show no new national MPI built from NFHS-6.

7.2 What did the Budget 2026-27 backgrounder say?

A PIB backgrounder of 13 February 2026 repeats the NITI Aayog series. The headcount ratio declined from 55.3 percent in 2005-06 to 14.96 percent in 2019-21. It then fell to 11.28 percent in 2022-23. It also puts extreme poverty at 5.3 percent in 2022-23.

7.3 How often do exams ask about the MPI?

The MPI was asked in UPSC CSP 2026, GS Paper I and in CDS I 2026, General Knowledge. It was also asked twice in CDS I 2023, General Knowledge. CDS I 2026 tested what the MPI measures: overlapping deprivations across health, education and living standards, not income or wealth. UPSC CSP 2026 tested the Alkire-Foster method, the 12 national indicators and which indicators are national only.

Four official questions are below, in the official wording. Read each key before you read the notes.

8.1 CDS I 2023, General Knowledge, Q67

CDS I 2023, General Knowledge, Q67.
Which one of the following indicators is included in the National Multidimensional Poverty Index (MPI)?
(a) Years of schooling
(b) Literacy rate
(c) Life expectancy
(d) Per capita income

Official paper: UPSC question paper. Official answer key (Series A): UPSC key. Key: (a).

The official Series A key gives (a) Years of schooling. Years of schooling is an Education indicator. Literacy rate, life expectancy and per capita income are not MPI indicators.

8.2 CDS I 2023, General Knowledge, Q96

CDS I 2023, General Knowledge, Q96.
According to the National Multi-dimensional Poverty Index (MPI) constructed by the NITI Aayog, a household is considered deprived if
1. a single member of the household is identified as under-nourished
2. the body mass index of a woman member, between 15 years and 49 years of age, is below 18.5 kg/m²
Select the correct answer using the code given below.
(a) 1 only
(b) 2 only
(c) Both 1 and 2
(d) Neither 1 nor 2

Official paper: UPSC question paper. Series A answer key from UPSC: key PDF. Key: (c).

The official Series A key gives (c) Both 1 and 2. Both statements describe the Nutrition indicator. One undernourished member marks the household deprived. A woman aged 15-49 is undernourished if her BMI is below 18.5. The Maternal Health indicator is a different rule.

8.3 CDS I 2026, General Knowledge, Q39

CDS I 2026, General Knowledge, Q39.
The Multidimensional Poverty Index (MPI) is measured by :
(a) Measuring each person’s overlapping deprivations across the indicators in the health, education and standard of living dimensions
(b) Measuring each person’s separate efficiencies across the indicators in the wealth, income and standard of living dimensions
(c) Measuring each person’s poverty on the indicators of health, happiness and ownership of vehicles
(d) Measuring inequalities and deprivations in respect of wealth, income and standard of living dimensions

Official paper: UPSC question paper. No public official answer key found. Site key: (a).

The site’s key is (a). No public official answer key was found for this paper. The answer follows UNDP’s wording: the MPI measures each person’s overlapping deprivations across health, education and standard of living. Option (b) talks about efficiencies. Option (c) names happiness and vehicles. Option (d) is about inequality.

8.4 UPSC CSP 2026, General Studies Paper I, Q100

UPSC CSP 2026, General Studies Paper I, Q100.
Consider the following statements about Multidimensional Poverty Index (MPI) :
1. MPI is calculated using Alkire-Foster methodology.
2. MPI calculated by NITI Aayog has a total of twelve indicators.
3. Maternal Health and Bank Account are common indicators in the MPI of NITI Aayog and MPI of United Nations Development Programme (UNDP).
Which of the statements given above is/are correct ?
(a) 1 and 2 only
(b) 1, 2 and 3
(c) 1 and 3 only
(d) 2 only

Official paper: UPSC question paper. No official key could be found online. Site key: (a).

The site’s key is (a) 1 and 2 only. Again, no public official key could be found. The sources support it. Statement 1 is true because both indices use the Alkire-Foster method. Statement 2 is true because the National MPI has 12 indicators. Statement 3 is false. Maternal Health and Bank Account are in India’s list only.

8.5 What traps should I avoid?

• Indicators: 12 national, 10 global. Maternal Health and Bank Account are national only.
• Baseline report (Nov 2021): NFHS-4, 2015-16. Progress review (2023): NFHS-5, 2019-21.
• BMI below 18.5 is the Nutrition rule, not the Maternal Health rule.
• 29.17% to 11.28% uses projected end years.
• Do not quote a Global MPI rank for India. UNDP does not rank countries.
• Do not mix 13.5 crore, 24.82 crore and 415 million.