Lesson 14: Multidimensional Poverty

Learning Objectives

After completing this lesson, students should be able to:

1. Define multidimensional poverty and explain its importance in development economics.
2. Explain why income alone is insufficient for measuring poverty.
3. Distinguish monetary poverty from multidimensional poverty.
4. Understand the capability approach and basic-needs approach.
5. Identify the major dimensions and indicators of multidimensional poverty.
6. Explain the Alkire-Foster methodology.
7. Understand deprivation cutoffs and the multidimensional poverty cutoff.
8. Calculate deprivation scores.
9. Calculate the multidimensional poverty headcount ratio.
10. Calculate poverty intensity.
11. Calculate the adjusted headcount ratio.
12. Explain the construction of a Multidimensional Poverty Index.
13. Understand the difference between censored and uncensored deprivation.
14. Analyze the strengths and limitations of multidimensional poverty measures.
15. Apply multidimensional poverty measures to development policy.

 

Poverty has traditionally been measured by looking at income or consumption. If a person’s income falls below a specified poverty line, that person or household is classified as poor.

This approach is extremely useful, but income is only one component of human well-being.

Imagine two households.

Household A has an income slightly above the poverty line but has:

• no access to clean drinking water,
• inadequate sanitation,
• poor housing,
• children who do not attend school,
• limited access to healthcare.

Household B has income below the monetary poverty line but has:

• access to clean water,
• adequate sanitation,
• good housing,
• regular school attendance,
• access to healthcare.

If poverty is measured exclusively through income, Household A may not be considered poor while Household B is classified as poor.

Yet Household A may experience serious disadvantages in several dimensions of life.

This problem provides the foundation for the concept of multidimensional poverty.

Multidimensional poverty recognizes that poverty is not simply a shortage of money. It can involve simultaneous deprivation in several aspects of human life.

2. Meaning of Multidimensional Poverty

Multidimensional poverty refers to poverty experienced through deprivation in multiple dimensions of well-being rather than solely through insufficient income or consumption.

These dimensions may include:

• Education
• Health
• Nutrition
• Housing
• Drinking water
• Sanitation
• Electricity
• Cooking fuel
• Employment
• Living standards
• Social participation
• Security
• Access to basic services

The exact dimensions depend on the poverty measure being used.

Therefore:Multidimensional Poverty≠Income Poverty AloneMultidimensional\ Poverty \neq Income\ Poverty\ Alone

A person can experience several forms of deprivation even when their monetary income does not place them below a conventional income poverty line.

3. Why Income Alone Is Not Enough

Income is an important determinant of well-being, but it does not automatically translate into capabilities.

Consider a household with relatively adequate income but poor access to healthcare because there is no healthcare facility nearby.

The household has financial resources but may still experience health deprivation.

Similarly, a family may have sufficient income but live in an area without:

• electricity,
• safe drinking water,
• sanitation,
• schools,
• roads.

Consequently, income provides command over resources, but development economics is also concerned with what people are actually able to do and be.

This distinction is central to multidimensional poverty analysis.

4. The Capability Approach

The intellectual foundation of multidimensional poverty is closely associated with the capability approach.

The capability approach emphasizes people’s:

• opportunities,
• freedoms,
• abilities,
• choices,
• functioning,
• access to basic conditions of life.

A person’s income is a means to achieving well-being, but it is not well-being itself.

For example:Income→Resources→Capabilities→FunctioningsIncome \rightarrow Resources \rightarrow Capabilities \rightarrow Functionings

Income can provide access to food, education and healthcare, but the final outcomes depend on many other factors.

Two people with identical incomes may have very different capabilities because of differences in:

• health,
• disability,
• location,
• public services,
• family responsibilities,
• social discrimination,
• infrastructure.

Therefore, multidimensional poverty attempts to capture deprivation more directly.

5. Poverty as Deprivation

Under a multidimensional approach, poverty is understood as deprivation.

A person may be deprived in:

Health

• Poor nutrition
• Limited healthcare
• High health risks

Education

• No schooling
• Children not attending school
• Low educational attainment

Living Standards

• Poor housing
• No electricity
• Unsafe drinking water
• Poor sanitation
• Inadequate cooking facilities

The important question becomes:

How many important dimensions of life does a person or household experience deprivation in, and how severe are those deprivations?

6. Monetary Poverty vs Multidimensional Poverty

The difference can be summarized as follows:

Monetary Poverty	Multidimensional Poverty
Focuses primarily on income or consumption	Focuses on several dimensions
Uses a monetary poverty line	Uses deprivation indicators and cutoffs
Measures economic resources	Measures broader deprivation
Easier to understand	More complex
Useful for cash-transfer policies	Useful for identifying multiple needs
May overlook non-monetary deprivation	Captures several forms of deprivation

Neither approach necessarily makes the other irrelevant.

Monetary poverty and multidimensional poverty answer different questions.

7. Example of Different Poverty Classifications

Consider two households.

Household A

Income is above the monetary poverty line.

However:

• Child does not attend school.
• Household lacks sanitation.
• Household lacks clean drinking water.
• Household has inadequate housing.

This household may not be income poor but may be multidimensionally poor.

Household B

Income is below the monetary poverty line.

However:

• Children attend school.
• Household has electricity.
• Safe drinking water is available.
• Housing is adequate.
• Healthcare is accessible.

This household may be income poor but experience fewer multidimensional deprivations.

This demonstrates why different poverty measures can identify different populations.

8. Dimensions of Multidimensional Poverty

A multidimensional poverty index generally contains several broad dimensions.

A common framework includes:

Dimension 1: Health

Possible indicators include:

• Nutrition
• Child mortality
• Access to healthcare

Dimension 2: Education

Possible indicators include:

• Years of schooling
• School attendance

Dimension 3: Living Standards

Possible indicators include:

• Electricity
• Drinking water
• Sanitation
• Housing
• Cooking fuel
• Assets

The specific dimensions and indicators depend on the methodology.

9. The Global Multidimensional Poverty Index Framework

A widely recognized global MPI framework uses three broad dimensions:

Health

• Nutrition
• Child mortality

Education

• Years of schooling
• School attendance

Living Standards

• Cooking fuel
• Sanitation
• Drinking water
• Electricity
• Housing
• Assets

The global framework uses weighted indicators to determine whether a household experiences sufficient deprivation to be considered multidimensionally poor.

National MPIs may use different dimensions, indicators, weights and thresholds according to national priorities and data availability.

10. Indicators and Deprivation Cutoffs

Each indicator requires a deprivation cutoff.

The deprivation cutoff determines when a person or household is considered deprived in a particular indicator.

For example:

Suppose an education indicator uses a minimum schooling threshold.

A household below the specified threshold is classified as deprived in education.

Similarly, a sanitation indicator may classify a household as deprived if it lacks access to adequate sanitation.

Thus:Indicator+Cutoff→Deprivation StatusIndicator + Cutoff \rightarrow Deprivation\ Status

The cutoff is therefore a critical methodological decision.

11. Binary Deprivation Indicators

Many multidimensional poverty measures initially represent deprivation using a binary value.

For example:dij={1if person/household i is deprived in indicator j0otherwised_{ij}= \begin{cases} 1 & \text{if person/household } i \text{ is deprived in indicator } j\\ 0 & \text{otherwise} \end{cases}

Thus:

• 1 = deprived
• 0 = not deprived

Suppose a household is deprived in:

• Education = 1
• Nutrition = 1
• Electricity = 0
• Sanitation = 1

Its deprivation vector is:(1,1,0,1)(1,1,0,1)

12. Weighting Indicators

Not every indicator must receive the same weight.

Weights reflect the relative importance assigned to different dimensions or indicators.

Suppose there are four indicators:

Indicator	Weight
Education	0.25
Health	0.25
Sanitation	0.25
Electricity	0.25

The total weight is:0.25+0.25+0.25+0.25=10.25+0.25+0.25+0.25=1

If a household is deprived in education, health and sanitation but not electricity:Score=0.25+0.25+0.25Score=0.25+0.25+0.25=0.75=0.75

The household therefore has a deprivation score of 0.75, or 75%.

13. Deprivation Score

The deprivation score measures the weighted proportion of indicators in which a household is deprived.

It can be represented as:ci=∑j=1dwjgijc_i=\sum_{j=1}^{d}w_jg_{ij}

Where:

• cic_i = deprivation score of household ii
• wjw_j = weight of indicator jj
• gijg_{ij} = deprivation status
• dd = number of indicators

The score usually lies between:0≤ci≤10 \leq c_i \leq 1

A score of 0 means no measured deprivation.

A score of 1 means deprivation in all measured indicators.

14. Multidimensional Poverty Cutoff

After calculating a deprivation score, we need a second threshold.

This is called the multidimensional poverty cutoff, commonly represented by:kk

The cutoff determines how many weighted deprivations are necessary for a household to be classified as multidimensionally poor.

For example, suppose:k=0.33k=0.33

Then a household with:ci≥0.33c_i\geq0.33

is classified as multidimensionally poor.

A household with:ci<0.33c_i<0.33

is not classified as multidimensionally poor under that particular measure.

15. Difference Between Deprivation Cutoff and Poverty Cutoff

Students frequently confuse these two concepts.

Deprivation cutoff

Determines whether a household is deprived in a particular indicator.

Multidimensional poverty cutoff

Determines whether the household has enough weighted deprivations to be classified as multidimensionally poor.

Therefore:Deprivation Cutoff→Indicator LevelDeprivation\ Cutoff \rightarrow Indicator\ Level

while:Poverty Cutoff→Overall Household LevelPoverty\ Cutoff \rightarrow Overall\ Household\ Level

16. The Alkire-Foster Method

One of the most influential approaches to multidimensional poverty measurement is the Alkire-Foster methodology.

It provides a systematic framework for:

1. Identifying who is poor.
2. Measuring how many people are poor.
3. Measuring the intensity of their poverty.

The methodology is particularly important because it can incorporate multiple dimensions and indicators while maintaining a clear identification rule.

17. Two Stages of the Alkire-Foster Method

The methodology can be understood through two main stages:

Stage 1: Identification

Determine who is multidimensionally poor.

Stage 2: Aggregation

Summarize multidimensional poverty across the population.

Thus:Identification→AggregationIdentification \rightarrow Aggregation

This two-stage structure is one of the central ideas of the methodology.

18. Step 1: Identify Deprivations

Suppose a household is evaluated using five indicators:

Indicator	Weight	Deprived?
Education	0.20	Yes
Health	0.20	Yes
Water	0.20	No
Sanitation	0.20	Yes
Electricity	0.20	No

The household’s deprivation score is:ci=(0.20)+(0.20)+(0.20)c_i=(0.20)+(0.20)+(0.20)ci=0.60c_i=0.60

Therefore, the household experiences 60% weighted deprivation.

19. Step 2: Apply the Poverty Cutoff

Suppose:k=0.40k=0.40

Since:0.60>0.400.60>0.40

the household is identified as multidimensionally poor.

Now consider another household with:ci=0.20c_i=0.20

Since:0.20<0.400.20<0.40

it is not classified as multidimensionally poor under this cutoff.

20. Step 3: Calculate the Headcount Ratio

Once multidimensionally poor households have been identified, the multidimensional poverty headcount ratio can be calculated.

It is:H=qnH=\frac{q}{n}

Where:

• qq = number of multidimensionally poor people
• nn = total population

Suppose:

• Total population = 1,000
• Multidimensionally poor population = 300

Then:H=3001000H=\frac{300}{1000}H=0.30H=0.30

or:H=30%H=30\%

Therefore, 30% of the population is multidimensionally poor.

21. Step 4: Calculate Poverty Intensity

The headcount ratio tells us how many people are poor.

But it does not tell us how many deprivations poor people experience on average.

For this we calculate poverty intensity.

It is represented by:AA

and can be expressed as:A=Total weighted deprivations among poor peopleTotal weighted possible deprivations among poor peopleA= \frac{\text{Total weighted deprivations among poor people}} {\text{Total weighted possible deprivations among poor people}}

In simplified terms, it represents the average proportion of weighted deprivations experienced by multidimensionally poor people.

22. Example of Poverty Intensity

Suppose there are four multidimensionally poor households.

Their deprivation scores are:

• Household A = 0.50
• Household B = 0.60
• Household C = 0.40
• Household D = 0.70

Then:A=0.50+0.60+0.40+0.704A= \frac{0.50+0.60+0.40+0.70}{4}A=2.204A=\frac{2.20}{4}A=0.55A=0.55

Therefore, the average deprivation intensity among the poor is:55%55\%

23. Adjusted Headcount Ratio

The adjusted headcount ratio is:M0=H×AM_0=H\times A

Where:

• M0M_0 = adjusted headcount ratio
• HH = multidimensional poverty headcount ratio
• AA = intensity of poverty

Suppose:H=0.30H=0.30

and:A=0.55A=0.55

Then:M0=0.30×0.55M_0=0.30\times0.55M0=0.165M_0=0.165

Therefore:M0=16.5%M_0=16.5\%

This means that adjusted multidimensional poverty is 16.5% under the specified measure.

24. Why Adjusted Headcount Is Better Than Headcount Alone

Suppose two countries have:HA=30%H_A=30\%

and:HB=30%H_B=30\%

Their poverty headcount is identical.

But suppose:AA=40%A_A=40\%

and:AB=70%A_B=70\%

Then:

Country A:M0=0.30×0.40=0.12M_0=0.30\times0.40=0.12M0=12%M_0=12\%

Country B:M0=0.30×0.70=0.21M_0=0.30\times0.70=0.21M0=21%M_0=21\%

Both countries have the same proportion of poor people, but poor people in Country B experience greater average deprivation.

This demonstrates the importance of measuring both incidence and intensity.

25. Complete Numerical Example

Consider a hypothetical economy with five households.

There are four equally weighted indicators:

• Education
• Health
• Sanitation
• Electricity

Each indicator receives:w=0.25w=0.25

Suppose the deprivation patterns are:

Household	Education	Health	Sanitation	Electricity
A	1	1	0	0
B	1	1	1	0
C	0	0	1	0
D	1	1	1	1
E	0	0	0	0

The deprivation scores are:

Household A

cA=0.25+0.25=0.50c_A=0.25+0.25=0.50

Household B

cB=0.25+0.25+0.25=0.75c_B=0.25+0.25+0.25=0.75

Household C

cC=0.25c_C=0.25

Household D

cD=1.00c_D=1.00

Household E

cE=0c_E=0

Suppose the poverty cutoff is:k=0.50k=0.50

Then:

• A → Poor
• B → Poor
• C → Not poor
• D → Poor
• E → Not poor

Therefore, 3 out of 5 households are multidimensionally poor.

The headcount ratio is:H=35=0.60H=\frac{3}{5}=0.60

Thus:H=60%H=60\%

Now calculate intensity among poor households:A=0.50+0.75+1.003A= \frac{0.50+0.75+1.00}{3}A=2.253A=\frac{2.25}{3}A=0.75A=0.75

Thus:A=75%A=75\%

The adjusted headcount ratio is:M0=H×AM_0=H\times AM0=0.60×0.75M_0=0.60\times0.75M0=0.45M_0=0.45

Therefore:M0=45%M_0=45\%

This example demonstrates the entire basic calculation.

26. Censored Deprivations

An important concept in multidimensional poverty measurement is censoring.

Suppose a person is deprived in an indicator but does not meet the overall poverty cutoff.

That deprivation may not be counted in the final measure of multidimensional poverty.

In contrast, when a person is identified as multidimensionally poor, their relevant deprivations are counted when calculating poverty intensity.

This distinction helps ensure that the measure focuses on deprivations experienced by people who have been identified as multidimensionally poor.

27. Uncensored vs Censored Headcounts

Uncensored headcount

Measures the proportion of the total population deprived in a particular indicator, regardless of whether those people are multidimensionally poor.

Censored headcount

Measures the proportion of the population that is both:

1. multidimensionally poor, and
2. deprived in that particular indicator.

The distinction is useful for policy.

Suppose 50% of the population lacks adequate sanitation, but only 30% are multidimensionally poor and deprived in sanitation.

The two statistics answer different questions.

28. Why Censored Indicators Matter for Policy

Suppose policymakers want to reduce multidimensional poverty.

They need to know not merely which deprivations are widespread, but which deprivations are concentrated among multidimensionally poor people.

For example:

• Education deprivation may be strongly concentrated among the poor.
• Electricity deprivation may be concentrated in rural areas.
• Sanitation deprivation may be widespread but particularly severe among poor households.

Censored indicators help identify the dimensions most directly associated with multidimensional poverty.

29. Dimensions and Indicators

It is important to distinguish between a dimension and an indicator.

Dimension

A broad area of human well-being.

Examples:

• Health
• Education
• Living standards

Indicator

A specific measurable variable within that dimension.

Examples:

Health:

• Nutrition
• Child mortality

Education:

• School attendance
• Years of schooling

Living standards:

• Electricity
• Water
• Sanitation

Therefore:Dimension→IndicatorsDimension \rightarrow Indicators

30. Equal and Unequal Weighting

There are two broad approaches to weighting.

Equal weighting

Every indicator receives the same weight.

For example, four indicators:w1=w2=w3=w4=0.25w_1=w_2=w_3=w_4=0.25

Unequal weighting

Some indicators receive greater importance.

For example:

Indicator	Weight
Health	0.30
Education	0.30
Housing	0.20
Electricity	0.10
Water	0.10

The total must equal:1.001.00

or 100%.

The choice of weights can substantially affect the measured level of poverty.

31. Choosing Poverty Cutoffs

The poverty cutoff kk is another major methodological decision.

Suppose:k=0.20k=0.20

Many households may be classified as poor.

If:k=0.50k=0.50

fewer households may qualify as multidimensionally poor.

If:k=0.70k=0.70

the measure identifies households experiencing very substantial deprivation.

Thus, changing kk changes the population identified as poor.

This is why researchers must justify their choice of cutoff.

32. Identification and Targeting

Multidimensional poverty measures can improve policy targeting.

Suppose a government knows only that a household is poor in monetary terms.

It may provide cash assistance.

But if multidimensional data reveal that the same household suffers from:

• poor sanitation,
• lack of electricity,
• school exclusion,
• inadequate housing,

then policy can become more targeted.

The government might combine:

• cash transfers,
• school support,
• sanitation programs,
• housing assistance,
• electricity connections.

Thus, multidimensional measurement can help policymakers move from income-based targeting toward deprivation-based targeting.

33. Multidimensional Poverty and Rural Areas

Multidimensional poverty is often closely related to geographical access to services.

Rural households may experience disadvantages because of:

• limited roads,
• lack of healthcare facilities,
• limited schools,
• inadequate sanitation,
• unreliable electricity,
• poor digital connectivity.

A household may have income but still face severe service-access deprivation.

Therefore, spatial and regional analysis is essential.

34. Multidimensional Poverty and Urban Areas

Urban poverty has a different structure.

Urban households may have greater physical access to:

• schools,
• hospitals,
• electricity,
• transportation.

However, they may face:

• overcrowded housing,
• insecure employment,
• high living costs,
• inadequate sanitation,
• informal settlements,
• housing insecurity.

Therefore, multidimensional poverty can occur in both rural and urban areas, although its characteristics may differ.

35. Multidimensional Poverty and Children

Children can experience poverty differently from adults.

For example, a child may suffer from:

• malnutrition,
• lack of schooling,
• unsafe housing,
• inadequate sanitation,
• limited healthcare.

Therefore, household income alone may not fully capture child deprivation.

Child-focused multidimensional poverty measures can provide additional information about the specific forms of deprivation experienced during childhood.

36. Multidimensional Poverty and Gender

Poverty can also have different consequences for men and women.

Women may face disadvantages related to:

• education,
• employment,
• income control,
• access to financial resources,
• healthcare,
• unpaid care responsibilities.

Household-level poverty measures may sometimes conceal differences between household members.

This creates an important measurement challenge:

A household may be classified as non-poor even though a particular individual within the household experiences significant deprivation.

Therefore, individual-level analysis can sometimes complement household-level measures.

37. Multidimensional Poverty and Inequality

Poverty and inequality are related but different concepts.

Poverty

Focuses on whether people experience deprivation below a specified threshold.

Inequality

Focuses on differences between people or groups.

A country may have:

• low multidimensional poverty but substantial inequality, or
• high multidimensional poverty with relatively similar levels of deprivation.

Therefore, both poverty and inequality should be examined separately.

38. Multidimensional Poverty and Economic Growth

Economic growth can reduce multidimensional poverty through:

• higher employment,
• higher incomes,
• increased government revenue,
• greater investment,
• improved infrastructure.

However, growth does not automatically eliminate multidimensional poverty.

Growth may occur alongside:

• regional disparities,
• poor education,
• inadequate healthcare,
• environmental degradation,
• unequal access to services.

Therefore:Economic Growth→Potential Poverty ReductionEconomic\ Growth \rightarrow Potential\ Poverty\ Reduction

but not necessarily:Economic Growth=Complete Poverty EliminationEconomic\ Growth = Complete\ Poverty\ Elimination

39. Multidimensional Poverty and Human Development

Human development and multidimensional poverty are closely related.

Human development asks:

What capabilities and opportunities do people possess?

Multidimensional poverty asks:

In which important dimensions are people experiencing severe deprivation?

Thus, multidimensional poverty analysis provides a way of identifying people who are deprived in important aspects of human development.

40. Multidimensional Poverty and the Sustainable Development Goals

Multidimensional poverty is relevant to several Sustainable Development Goals.

It is particularly connected with:

• poverty reduction,
• hunger and nutrition,
• health,
• education,
• gender equality,
• clean water and sanitation,
• energy,
• decent work,
• reduced inequalities.

This demonstrates that poverty reduction cannot be separated from broader development objectives.

41. Advantages of Multidimensional Poverty Measurement

1. Broader Understanding of Poverty

It recognizes that poverty has several dimensions.

2. Identifies Multiple Deprivations

It shows which specific disadvantages people experience.

3. Better Policy Targeting

It can help governments design interventions around actual deprivations.

4. Captures Non-Monetary Poverty

Important services and capabilities may be captured even when income is not.

5. Measures Intensity

It can distinguish between people experiencing different numbers or degrees of deprivation.

6. Supports Regional Analysis

It can identify differences between rural and urban areas or different regions.

7. Supports Policy Coordination

Education, health, housing and infrastructure policies can be considered together.

42. Limitations of Multidimensional Poverty

Despite its advantages, multidimensional poverty measurement has important limitations.

42.1 Indicator Selection

Researchers must decide which dimensions to include.

Different choices can produce different results.

42.2 Weighting Problems

How much importance should be given to health compared with education or housing?

There is no universally uncontested answer.

42.3 Cutoff Selection

Results can change when deprivation thresholds change.

42.4 Poverty Cutoff

The choice of kk affects how many households are classified as poor.

42.5 Data Requirements

Multidimensional measures require detailed household-level information.

42.6 Data Quality

Incomplete or inaccurate survey data can affect results.

42.7 Comparability

National MPIs may use different dimensions, indicators and thresholds, making comparisons difficult.

42.8 Household-Level Measurement

Household-level data can hide differences between individuals.

42.9 Correlation Between Indicators

Some deprivations may be strongly correlated.

For example, poor housing may be associated with poor sanitation and lack of electricity.

42.10 Oversimplification

Reducing complex human experiences to a single index can hide important details.

---

43. The Problem of Weighting

Weighting is one of the most important methodological issues.

Suppose:Health=0.50Health=0.50Education=0.20Education=0.20Housing=0.30Housing=0.30

The result depends partly on how these dimensions have been weighted.

A different weighting system could produce a different poverty classification.

Therefore, researchers should conduct sensitivity analysis to determine whether conclusions remain similar under reasonable alternative assumptions.

44. Sensitivity Analysis

Suppose a country has a multidimensional poverty rate of:25%25\%

under one set of weights.

Under another reasonable weighting system:27%27\%

Under another:24%24\%

The broad conclusion may be relatively stable.

However, if the results change from:10%10\%

to:50%50\%

depending on the weighting system, the measurement is highly sensitive to methodological choices.

Sensitivity analysis is therefore important in serious poverty research.

45. Data Requirements

Multidimensional poverty measurement generally requires household or individual data covering several dimensions.

A useful survey may contain information about:

• Household composition
• Education
• School attendance
• Health
• Nutrition
• Housing
• Water
• Sanitation
• Electricity
• Assets
• Employment
• Income or consumption

The quality of the poverty index depends heavily on the quality of the underlying data.

46. National MPI vs Global MPI

A global MPI is designed to provide a broadly comparable measure across countries using a common methodology.

A national MPI is designed for a particular country and can incorporate nationally relevant dimensions and indicators.

For example, a country may consider:

• housing quality,
• employment security,
• social protection,
• financial inclusion,
• access to transport,

in addition to the dimensions used in a global measure.

Therefore, national and global MPIs serve somewhat different purposes.

47. Multidimensional Poverty as a Policy Dashboard

An MPI should not be viewed only as a ranking number.

Its greater value may come from identifying the specific deprivations contributing to poverty.

Suppose:MPI=0.20MPI=0.20

That number alone tells us relatively little.

But suppose decomposition reveals:

• 35% of deprivation comes from education,
• 25% from health,
• 20% from sanitation,
• 10% from housing,
• 10% from electricity.

Policymakers now have information about where interventions may be needed.

This is one of the most important practical advantages of multidimensional poverty measurement.

48. Decomposition of Multidimensional Poverty

Multidimensional poverty measures can often be decomposed by:

• Region
• Gender
• Age
• Rural/urban location
• Ethnic or social groups where appropriate
• Dimensions
• Individual indicators

For example:National MPI=Rural MPI+Urban MPINational\ MPI = Rural\ MPI + Urban\ MPI

conceptually, with appropriate population weighting.

Such decomposition can identify where deprivation is concentrated.

49. Practical Policy Example

Imagine a hypothetical country called Developmentia.

The government discovers:

• 28% of people are multidimensionally poor.
• 20% experience educational deprivation.
• 25% experience sanitation deprivation.
• 15% experience electricity deprivation.
• 10% experience nutrition deprivation.

The government could design a coordinated strategy:

Education

• Expand schools
• Provide scholarships
• Improve attendance

Sanitation

• Build sanitation infrastructure
• Improve waste management
• Expand clean-water systems

Electricity

• Expand grid connections
• Support off-grid systems

Nutrition

• Improve maternal and child nutrition programs
• Strengthen food security

The MPI therefore becomes a policy diagnostic tool rather than merely a statistic.

50. A Second Numerical Example

Suppose a country has 2,000 households.

After applying the chosen indicators and poverty cutoff:q=600q=600

households are identified as multidimensionally poor.

Then:H=6002000H=\frac{600}{2000}H=0.30H=0.30

Therefore:H=30%H=30\%

Suppose the average weighted deprivation among poor households is:A=0.60A=0.60

Then:M0=H×AM_0=H\times AM0=0.30×0.60M_0=0.30\times0.60M0=0.18M_0=0.18

Thus:M0=18%M_0=18\%

The interpretation is:

• 30% of households are identified as multidimensionally poor.
• The average poor household experiences 60% of the weighted deprivations represented in the measure.
• The adjusted headcount ratio is 18%.

51. Important Formulas

Multidimensional Headcount Ratio

H=qnH=\frac{q}{n}

Deprivation Score

ci=∑jwjgijc_i=\sum_j w_jg_{ij}

Poverty Identification Rule

ci≥kc_i\geq k

Poverty Intensity

A=Total weighted deprivations among poorTotal possible weighted deprivations among poorA= \frac{\text{Total weighted deprivations among poor}} {\text{Total possible weighted deprivations among poor}}

Adjusted Headcount Ratio

M0=H×AM_0=H\times A

These formulas are particularly important for examinations.

52. Important Conceptual Distinctions

Students should remember the following distinctions.

Income poverty

Insufficient monetary resources.

Multidimensional poverty

Deprivation across several dimensions.

Deprivation cutoff

Determines whether a household is deprived in a particular indicator.

Poverty cutoff

Determines whether overall deprivation is sufficient to classify the household as multidimensionally poor.

Headcount ratio

Measures how many people are poor.

Intensity

Measures how much deprivation poor people experience on average.

Adjusted headcount ratio

Combines incidence and intensity.

53. Common Misconceptions

Misconception 1

“Multidimensional poverty means ignoring income.”

Incorrect.

Income remains important. Multidimensional poverty simply recognizes that income is not the only relevant dimension.

---

Misconception 2

“Anyone deprived in one indicator is multidimensionally poor.”

Incorrect.

The classification normally depends on an overall poverty cutoff.

---

Misconception 3

“All MPIs use exactly the same indicators.”

Incorrect.

Different national and international measures may use different dimensions, indicators and thresholds.

---

Misconception 4

“A higher poverty headcount always means greater multidimensional poverty.”

Not necessarily.

Intensity must also be considered.

---

Misconception 5

“One multidimensional poverty index is universally correct.”

Incorrect.

Measurement involves methodological choices regarding dimensions, indicators, weights and cutoffs.

54. Relationship Between H, A and M₀

This relationship is fundamental:M0=H×AM_0=H\times A

Where:H=IncidenceH=\text{Incidence}

and:A=IntensityA=\text{Intensity}

Therefore:Multidimensional Poverty=Incidence×IntensityMultidimensional\ Poverty = Incidence\times Intensity

This provides a much richer understanding than simply counting poor people.

55. Interpretation of Changes in MPI

Suppose:

Year 1:H=40%,A=50%H=40\%, \quad A=50\%

Then:M0=0.40×0.50=0.20M_0=0.40\times0.50=0.20

or:20%20\%

In Year 2:H=30%,A=45%H=30\%, \quad A=45\%

Then:M0=0.30×0.45=0.135M_0=0.30\times0.45=0.135

or:13.5%13.5\%

The decline could reflect:

• fewer people becoming multidimensionally poor,
• reduced intensity among poor people,
• or both.

Therefore, policymakers should examine both HH and AA, not only M0M_0.

56. Why Multidimensional Poverty Matters for Development Economics

Development economics is fundamentally concerned with improving human well-being and expanding people’s opportunities.

Multidimensional poverty is important because it recognizes that deprivation can take many forms.

A person may have income but lack:

• education,
• healthcare,
• clean water,
• sanitation,
• adequate housing,
• electricity,
• nutrition,
• security.

Therefore, development policy must often address several constraints simultaneously.

This leads to a central development principle:Poverty Reduction≠Income Increase AlonePoverty\ Reduction \neq Income\ Increase\ Alone

Instead:Poverty Reduction=Income+Human Capabilities+Basic Services+OpportunitiesPoverty\ Reduction = Income + Human\ Capabilities + Basic\ Services + Opportunities

57. Summary

Multidimensional poverty expands the traditional understanding of poverty beyond income and consumption.

The approach recognizes that human deprivation can occur in several dimensions, including health, education and living standards.

The capability approach provides an important conceptual foundation by emphasizing people’s freedoms and capabilities rather than income alone.

The Alkire-Foster methodology provides a systematic framework for identifying multidimensionally poor people and aggregating their deprivation.

The process involves:

1. Selecting dimensions.
2. Selecting indicators.
3. Establishing deprivation cutoffs.
4. Assigning weights.
5. Calculating deprivation scores.
6. Applying a multidimensional poverty cutoff.
7. Identifying poor people.
8. Calculating the headcount ratio.
9. Calculating poverty intensity.
10. Calculating the adjusted headcount ratio.

The key formula is:M0=H×AM_0=H\times A

where HH represents the incidence of multidimensional poverty and AA represents the average intensity of deprivation among the poor.

The major advantage of multidimensional poverty measurement is that it reveals what people are deprived of, not merely how much money they have.

At the same time, multidimensional measures involve important methodological choices concerning dimensions, indicators, weights, thresholds and data.

Key Terms

• Multidimensional Poverty
• Monetary Poverty
• Capability Approach
• Functionings
• Deprivation
• Deprivation Cutoff
• Poverty Cutoff
• Alkire-Foster Method
• Deprivation Score
• Headcount Ratio
• Poverty Intensity
• Adjusted Headcount Ratio
• Censored Headcount
• Uncensored Headcount
• Multidimensional Poverty Index
• Indicator
• Dimension
• Weight
• Identification
• Aggregation
• Sensitivity Analysis
• Human Development
• Basic Needs
• Poverty Decomposition

Revision Points

1. Poverty is multidimensional.
2. Income is an important but incomplete measure of well-being.
3. Multidimensional poverty focuses on multiple forms of deprivation.
4. The capability approach emphasizes opportunities and freedoms.
5. Health, education and living standards are common dimensions.
6. Each indicator requires a deprivation cutoff.
7. Indicators can receive different weights.
8. A deprivation score summarizes a household’s weighted deprivations.
9. The poverty cutoff determines who is multidimensionally poor.
10. The Alkire-Foster methodology separates identification from aggregation.
11. The headcount ratio measures the incidence of poverty.
12. Poverty intensity measures the average deprivation among poor people.
13. The adjusted headcount ratio combines incidence and intensity.
14. The basic relationship is M0=H×AM_0=H\times A.
15. Censored and uncensored headcounts answer different questions.
16. Multidimensional poverty can be decomposed by dimensions and population groups.
17. National and global MPIs may use different methodologies.
18. Weighting and cutoff choices can affect results.
19. Multidimensional poverty measurement requires detailed data.
20. MPI analysis can help governments target specific deprivations.

Short-Answer Questions

1. Define multidimensional poverty.
2. Why is income alone insufficient for measuring poverty?
3. What is the capability approach?
4. What is a deprivation cutoff?
5. What is a deprivation score?
6. What is the multidimensional poverty cutoff?
7. What is the Alkire-Foster methodology?
8. Define the multidimensional poverty headcount ratio.
9. What is poverty intensity?
10. Define the adjusted headcount ratio.
11. What is meant by censored deprivation?
12. What is an uncensored headcount?
13. Distinguish between a dimension and an indicator.
14. Why are weights used in an MPI?
15. What is sensitivity analysis?
16. How does multidimensional poverty differ from monetary poverty?
17. Why is sanitation an important poverty indicator?
18. How can MPI data assist government policy?
19. What is the difference between incidence and intensity?
20. Why can multidimensional poverty exist even when income poverty is low?

Long-Answer Questions

1. Define multidimensional poverty and explain its importance in modern development economics.
2. Why is income alone insufficient to measure poverty? Discuss.
3. Explain the capability approach as a foundation of multidimensional poverty analysis.
4. Compare monetary poverty with multidimensional poverty.
5. Explain the major dimensions and indicators commonly used in multidimensional poverty measurement.
6. Explain the Alkire-Foster methodology in detail.
7. Explain the concepts of deprivation cutoff and multidimensional poverty cutoff.
8. Describe how a deprivation score is calculated.
9. Explain the multidimensional poverty headcount ratio and poverty intensity.
10. Derive and explain the adjusted headcount ratio.
11. Discuss the advantages of multidimensional poverty measurement.
12. Discuss the limitations of the Multidimensional Poverty Index.
13. Explain censored and uncensored deprivation.
14. Discuss the importance of weighting in multidimensional poverty measurement.
15. Explain how multidimensional poverty can be decomposed by population groups and indicators.
16. Discuss the relationship between multidimensional poverty and human development.
17. Explain how multidimensional poverty measurement can improve policy targeting.
18. Discuss rural and urban dimensions of multidimensional poverty.
19. Explain the relationship between multidimensional poverty, inequality and economic growth.
20. “Poverty is more than a shortage of income.” Discuss the statement using the multidimensional approach.

Multiple-Choice Questions

1. Multidimensional poverty focuses primarily on:

A. Income alone
B. Multiple forms of deprivation
C. GDP growth alone
D. Exports only

Answer: B. Multiple forms of deprivation

2. Which approach strongly influenced modern multidimensional poverty analysis?

A. Capability approach
B. Quantity theory of money
C. Comparative advantage theory
D. Classical population theory

Answer: A. Capability approach

3. A deprivation cutoff determines:

A. Whether a country is developed
B. Whether a household is deprived in a particular indicator
C. The exchange rate
D. GDP growth

Answer: B. Whether a household is deprived in a particular indicator

4. The poverty cutoff kk determines:

A. The inflation rate
B. Whether overall weighted deprivation is sufficient for multidimensional poverty classification
C. The unemployment rate
D. The GDP growth rate

Answer: B. Whether overall weighted deprivation is sufficient for multidimensional poverty classification

5. The Alkire-Foster method contains which two broad stages?

A. Production and consumption
B. Identification and aggregation
C. Saving and investment
D. Growth and inflation

Answer: B. Identification and aggregation

6. The multidimensional poverty headcount ratio is:

A. q/nq/n
B. n/qn/q
C. H+AH+A
D. H−AH-A

Answer: A. q/nq/n

7. Poverty intensity measures:

A. Total GDP
B. Average deprivation among multidimensionally poor people
C. Total population
D. Inflation

Answer: B. Average deprivation among multidimensionally poor people

8. The adjusted headcount ratio is:

A. H+AH+A
B. H−AH-A
C. H×AH\times A
D. H/AH/A

Answer: C. H×AH\times A

9. Which is commonly included as a living-standard indicator?

A. Electricity
B. Stock-market index
C. Interest rate
D. Exchange rate

Answer: A. Electricity

10. Which of the following is an education indicator?

A. Years of schooling
B. Inflation
C. Exports
D. Exchange rate

Answer: A. Years of schooling

11. If H=0.40H=0.40 and A=0.50A=0.50, then M0M_0 equals:

A. 0.10
B. 0.20
C. 0.45
D. 0.90

Answer: B. 0.20

12. If a household has a deprivation score below the poverty cutoff, it is:

A. Automatically extremely poor
B. Not classified as multidimensionally poor under that cutoff
C. Automatically income poor
D. Automatically unemployed

Answer: B. Not classified as multidimensionally poor under that cutoff

13. Which statement is correct?

A. All MPIs use identical indicators.
B. Multidimensional poverty ignores education.
C. Different poverty measures can use different dimensions and thresholds.
D. Income is irrelevant to poverty.

Answer: C. Different poverty measures can use different dimensions and thresholds.

14. Censored headcounts focus on:

A. All people deprived in an indicator
B. People who are both multidimensionally poor and deprived in the indicator
C. Only rich households
D. Only employed people

Answer: B. People who are both multidimensionally poor and deprived in the indicator

15. A major advantage of multidimensional poverty measurement is that it:

A. Eliminates all measurement problems
B. Identifies multiple forms of deprivation
C. Requires no data
D. Measures GDP directly

Answer: B. Identifies multiple forms of deprivation

Final Examination Framework

A comprehensive examination question could be:

“Define multidimensional poverty and explain why income alone is insufficient to measure poverty. Discuss the capability approach and explain the Alkire-Foster methodology, including deprivation cutoffs, poverty cutoffs, headcount ratio, poverty intensity and the adjusted headcount ratio. Use a numerical example to illustrate the measurement process.”

A strong answer should contain:

1. Definition of multidimensional poverty.
2. Limitations of monetary poverty.
3. Capability approach.
4. Dimensions and indicators.
5. Deprivation cutoffs.
6. Indicator weights.
7. Deprivation score.
8. Poverty cutoff.
9. Identification of multidimensionally poor people.
10. Headcount ratio.
11. Poverty intensity.
12. Adjusted headcount ratio.
13. Numerical example.
14. Advantages.
15. Limitations.
16. Policy applications.
17. Conclusion.

Final Takeaway

The central contribution of multidimensional poverty analysis is that it changes the question from:

“How much money does a person have?”

to a broader question:

“What important things is this person deprived of, and how many of those deprivations are experienced simultaneously?”

Income remains an important dimension of economic well-being, but development economics recognizes that people need much more than income to live productive and dignified lives.

Education, health, nutrition, housing, sanitation, clean water, electricity and access to basic services all influence human capabilities.

The most important relationship to remember is:M0=H×A\boxed{M_0=H\times A}

where:H=incidence of multidimensional povertyH=\text{incidence of multidimensional poverty}

and:A=average intensity of deprivation among the multidimensionally poorA=\text{average intensity of deprivation among the multidimensionally poor}

Thus, multidimensional poverty measurement provides a richer understanding of poverty by combining who is poor with how intensely they are deprived.

This makes it one of the most important modern tools for understanding poverty and designing development policies.