Chapter 5

Poverty, Inequality,
and Development
Distribution and Development: Eight Critical Questions
• How can we best measure inequality and poverty?

• What is the extent of relative inequality in developing countries; how is this related to
the extent of poverty?

• Who are the poor, and what are their economic characteristics?

• What determines the nature of economic growth—that is, who benefits from
economic growth, and why?

• Are rapid economic growth and more equal income distribution compatible or
conflicting objectives? Is rapid growth achievable only at a cost of greater income
inequality or can lessening income disparities contribute to higher growth rates?

• Do the poor benefit from growth, and does this depend on the type of growth a
developing country experiences? What might be done to help the poor benefit more?

• What is so bad about extreme inequality?

• What kinds of policies are required to reduce the magnitude and extent of absolute
poverty?
Measuring Inequality
• 1. The Personal or Size Distributions
 The distribution of income according to size class without regard to the sources of
that income.
 Quintiles (every 20%)
 Deciles (every 10%)
Measuring Inequality
• 1. The Personal or Size Distributions
• 2. Lorenz Curve
 A graph depicting the variance of the size distribution of income from perfect
equality.
Measuring Inequality
• 1. The Personal or Size Distributions
• 2. Lorenz Curve
• 3. Gini Coefficients and Aggregate Measures of Inequality
 An aggregate numerical measure of income inequality ranging from 0 (perfect
equality) to 1 (perfect inequality). It is measured graphically by dividing the area
between the perfect equality line and the Lorenz curve by the total area lying to the
right of the equality line in a Lorenz diagram
Measuring Inequality
• 1. The Personal or Size Distributions
• 2. Lorenz Curve
• 3. Gini Coefficients and Aggregate Measures of Inequality
• 4. The Functional or Factor Share Distribution of Income
 The distribution of income to factors of distribution without regard to the
ownership of the factors.
 Factors of Production are resources or inputs required to produce a good or a
service, such as land, labor, and capital.
Desirable Properties for Inequality Measures
• 1. Anonymity
 measure should not depend on who has higher income; e.g. whether we believe the
rich or poor to be good or bad people

• 2. Scale Independence
 measure should not depend on the size of the economy

• 3. Population Independence
 measure should not based on the number of income recipients

• 4. Transfer Principle
 all other incomes constant, if transfer income from a richer to a poorer, the resulting
new income distribution is more equal

• Gini coefficient satisfies all four properties and is widely used as a measure of a
nation’s extent of inequality.
Measuring Absolute Poverty
• Absolute Poverty is a situation of being unable or only barely able to meet
the subsistence essentials of food, clothing, and shelter.
• Headcount, 𝐻, is the number of those whose incomes fall below the absolute
poverty line, 𝑌𝑝 (usually defined as living on less than $1.25 a day or $2 a day
in PPP dollars).
• Headcount Index (or Headcount Ratio) is the headcount as a fraction of total
population, 𝐻/𝑁, that is, the proportion of a country’s population living
below the poverty line.
• Total Poverty Gap (TPG) is the sum of of the difference between the poverty
line and actual income levels of all people living below that line. It tells us the
total amount of income necessary to raise everyone who is below the poverty
line up to that line.
 𝑻𝑷𝑮 = σ𝑯
𝒊=𝟏(𝒀𝒑 − 𝒀𝒊 )
Measuring Absolute Poverty (cont.)
𝑇𝑃𝐺
• Average Poverty Gap, 𝐴𝑃𝐺 =
𝑁
𝐴𝑃𝐺
• Normalized Poverty Gap, 𝑁𝑃𝐺 = , measures the the size of average poverty
𝑌𝑝
gap in relation to the poverty line. NPG is between 0 and 1.
𝑇𝑃𝐺
• Average Income Shortfall, 𝐴𝐼𝑆 = , tells us the average amount by which the
𝐻
income of a poor person falls below the poverty line.
𝐴𝐼𝑆
• Normalized Income Shortfall, 𝑁𝐼𝑆 =
𝑌𝑝
Desirable Properties of Poverty Measure
• 1. Anonymity
• 2. Population Independence
• 3. Monotonicity
 if you add income to someone below the poverty line, all other incomes held
constant, poverty can be no greater than it was

• 4. Distributional Sensitivity
 other things being equal, if you transfer income from a poor person to a richer
person, the resulting outcome will be poorer

• Headcount ratio satisfies 1, 2, and 3, but fails 4; simple headcount also fails 2;
TPG fails 2
The Foster-Greer-Thorbecke (FGT) Index
• The FGT index is a class of measures of the level of absolute poverty
𝛼
1 𝐻 𝑌 −𝑌
• 𝑃𝛼 = σ𝑖=1 𝑝 𝑖
𝑁 𝑌𝑝

• 𝛼=0
 𝑃0 = 𝐻/𝑁, we get the headcount ratio

• 𝛼=1
 𝑃1 = 𝑁𝑃𝐺
𝐻
 𝑃1 = ∗ 𝑁𝐼𝑆
𝑁
 Poverty goes up whenever either the fraction of people in poverty increases or
poverty depth goes up (better than 𝑃0 , but still not satisfy distributional
sensitivity)
• 𝛼=2
2
1 𝐻 𝑌 −𝑌
 𝑃2 = σ𝑖=1 𝑝 𝑖
𝑁 𝑌𝑝

 raising the income for a household living at 50% of the poverty line will generate
an impact five times greater than raising by the same amount for a household
living at 90% of the poverty line (distributional sensitivity)
 captures the severity of poverty
 At 𝑃0 , headcount ratio creates incentives for officials to focus efforts on the ones
closest to the poverty line (easiest). But at 𝑃2 , officials should address the poorest
of the poor (another reason 𝑃2 is preferred to 𝑃0 )
Poverty: Income or Multidimensional

• If we must have a single indicator for poverty, income measurements, like 𝑃2 ,

satisfy all desirable properties.
• But in general is measuring income sufficient?
• Recall Sen’s capacity framework
• Poverty must be conceptualized – and so measured – in a multidimensional
way.
• Newly introduced Multidimensional Poverty Index (MPI)
What Is So Bad about Extreme Inequality?
• 1. Extreme income inequality leads to economic inefficiency.
 Lack of collateral
 Cannot borrow to educate children or start a business
 The saving rate in the economy is low

• 2. Extreme income inequality undermines social stability and solidarity.

 The rich have greater political power and economic bargaining power.
 They can use these powers to create outcomes that are favorable to them.
 These outcomes may be diverted from the purposes of fast economic growth.

• 3. Extreme income inequality is generally viewed as unfair.

• 𝑊 = 𝑊(𝑌, 𝐼, 𝑃), where W is social welfare, Y is income, I is inequality, P is poverty
• +, -, -
The Dualistic-Development Thesis
• Dualism
 This is a world of dual societies, of rich nations and poor nations.
 The divergences are existent, persistent, and even increasing.
 Coexistence of “superior” and “inferior” (rural agriculture sector vs urban
industrial sector; educated rich vs illiterate poor; wealthy industrialized countries vs
impoverished peasant societies)
 This coexistence is chronic and not merely transitional.
 The superior do little or nothing to pull up the inferior.
How Lorenz Curves Shift in Some Cases of Dualistic Development?

• 1. The traditional-sector enrichment growth topology

 all benefits of growth are divided among traditional-sector workers, with little or no
growth occurring in the modern sector
 focuses on the substantial reduction of absolute poverty
 low income levels and low growth rates
 examples: Sri Lanka, some states in the west of India
Improved Income Distribution under the Traditional-
Sector Enrichment Topology
How Lorenz Curves Shift in Some Cases of Dualistic Development?
• 1. The traditional-sector enrichment growth topology
• 2. The modern-sector enrichment growth topology

 growth benefits only a fixed number of people in the modern sector

 hold constant both the number of workers and wages in the traditional sector
 examples: Latin America and Africa
Worsened Income Distribution under the Modern-
Sector Enrichment Topology
How Lorenz Curves Shift in Some Cases of Dualistic Development?
• 1. The traditional-sector enrichment growth topology
• 2. The modern-sector enrichment growth topology
• 3. The modern-sector enlargement growth topology

 enlarge the size of the modern sector while maintaining constant wages in both sectors
 examples: western developed countries, and some East Asian economies (China, South
Korea, Taiwan)
Crossing Lorenz Curves in the Modern-Sector
Enlargement Growth Topology
Simon Kuznets’ Inverted-U Hypothesis
Selected Income Distribution Estimates
Kuznets Curve with Latin America Countries Identified
Growth and Inequality
• Is there any relationship between economic growth and inequality?
• Growth rate is one thing.
• The character of economic growth is another.
 how it is achieved?
 who participates?
 which sectors are given priority?
 what institutional arrangements are designed and emphasized?

• The degree to which economic growth improves living standards of the

poor
• High growth is not necessarily sustained by increasing inequality.
Absolute Poverty: Extent and Magnitude
• Progress on Extreme Poverty

 clear progress on $1.25-a-day headcount

o 1.94 billion in the world in 1981
o 1.22 billion in 2010
o 37% reduction despite 53% population growth in the period

 less clear progress on a $2.00-per-day headcount

o 8% reduction from 1981 to 2010
o partly due to those crossing the $1.25 line but remaining under $2 line
The Incidence of Extreme Poverty is Uneven
Multidimensional Poverty Index (MPI)

• Identification of poverty status through a dual cutoff:

• First, cutoff levels within each dimension (analogous to falling below a
poverty line for example $1.25 per day for income poverty);
• Second, cutoff in the number of dimensions in which a person must be
deprived (below a line) to be deemed multidimensionally poor.
• MPI focuses on deprivations in health, education, and standard of living;
and each receives equal (that is one-third of the overall total) weight, 1/3.
MPI Indicators
• 1. Health
 two indicators with equal weight
 each accounts as 1/6 toward the maximum deprivation in the MPI
o whether any child has died in the family
o whether any adult or child in the family is malnourished

• 2. Education
• 3. Standard of Living
MPI Indicators
• 1. Health
• 2. Education
 two indicators with equal weight
 each accounts as 1/6 toward the MPI
o whether no household member completed 5 years of schooling
o whether any school-aged child is out of school for grades 1 through 8

• 3. Standard of Living
MPI Indicators
• 1. Health
• 2. Education
• 3. Standard of Living
 equal weight on 6 deprivations
 each accounts as 1/18 toward the maximum
o lack of electricity
o insufficiently safe drinking water
o inadequate sanitation
o inadequate flooring
o unimproved cooking fuel
o lack of more than one of 5 assets – telephone, radio, TV, bicycle, and motorbike
MPI Indicators
• 1. Health
• 2. Education
• 3. Standard of Living
• Individuals are identified as “multidimensionally poor” when their family is
deprived by a ”weighted sum” of 0.333 or more (3 out of 10 points as
calculated)
Indicator
Education
No one has completed 5 years of schooling
At least one school-aged child not enrolled in school
Health
At least one member is malnourished
One or more children have died
Living Standards
No electricity
No access to clean drinking water
No access to adequate sanitation
House has dirty floor
Household uses dirty cooking fuel (e.g. charcoal)
Household lacks more than one of the 5 assets
Score
Multidimensionally poor (>= 0.333)?
Household Weight
1 2 3
0 1/6 = 0.167
0 1/6 = 0.167
0 1/6 = 0.167
1 1/6 = 0.167
0 1/18 = 0.056
0 1/18 = 0.056
0 1/18 = 0.056
0 1/18 = 0.056
1 1/18 = 0.056
0 1/18 = 0.056
0.223
No
Indicator Household Weight
1 2 3
Education
No one has completed 5 years of schooling 1 1/6 = 0.167
At least one school-aged child not enrolled in school 1 1/6 = 0.167
Health
At least one member is malnourished 0 1/6 = 0.167
One or more children have died 1 1/6 = 0.167
Living Standards
No electricity 1 1/18 = 0.056
No access to clean drinking water 0 1/18 = 0.056
No access to adequate sanitation 1 1/18 = 0.056
House has dirty floor 0 1/18 = 0.056
Household uses dirty cooking fuel (e.g. charcoal) 1 1/18 = 0.056
Household lacks more than one of the 5 assets 1 1/18 = 0.056
Score 0.725
Multidimensionally poor (>= 0.333)? Yes
Indicator Household Weight
1 2 3
Education
No one has completed 5 years of schooling 1 1/6 = 0.167
At least one school-aged child not enrolled in school 0 1/6 = 0.167
Health
At least one member is malnourished 0 1/6 = 0.167
One or more children have died 1 1/6 = 0.167
Living Standards
No electricity 1 1/18 = 0.056
No access to clean drinking water 0 1/18 = 0.056
No access to adequate sanitation 0 1/18 = 0.056
House has dirty floor 0 1/18 = 0.056
Household uses dirty cooking fuel (e.g. charcoal) 1 1/18 = 0.056
Household lacks more than one of the 5 assets 1 1/18 = 0.056
Score ?
Multidimensionally poor (>= 0.333)? ?
Indicator Household Weight
1 2 3
Education
No one has completed 5 years of schooling 1 1/6 = 0.167
At least one school-aged child not enrolled in school 0 1/6 = 0.167
Health
At least one member is malnourished 0 1/6 = 0.167
One or more children have died 1 1/6 = 0.167
Living Standards
No electricity 1 1/18 = 0.056
No access to clean drinking water 0 1/18 = 0.056
No access to adequate sanitation 0 1/18 = 0.056
House has dirty floor 0 1/18 = 0.056
Household uses dirty cooking fuel (e.g. charcoal) 1 1/18 = 0.056
Household lacks more than one of the 5 assets 1 1/18 = 0.056
Score 0.502
Multidimensionally poor (>= 0.333)? Yes
Interaction of the Deprivations

• Building the index from household measures up to the aggregate measure

(rather than using already-aggregated statistics), MPI approach takes account
of multiplied or interactive harm (complementarity) done when multiple
deprivations are experienced by the same individual or family.
• The MPI approach assumes an individual’s lack of capability in one area can
only to a degree be made up by other capabilities – capabilities are treated as
substitutes up to a point but then as complements.
Compute the MPI for A Country
• The actual MPI for the country (or region or group) is computed with the
adjusted headcount ratio
• 𝐻𝑀 ∗ 𝐴
• where 𝐻𝑀 is the headcount ratio (percentage of people living in
multidimensional poverty); 𝐴 is the average intensity of deprivation.
• The adjusted headcount ratio satisfies all four desirable properties.
• Important example –
 dimensional monotonicity: when a person deemed poor becomes deprived in
another indicator, s/he is deemed even poorer.
MPI Tells A Different Story than Income

• The results showed that knowing income poverty is not enough if our
concern is with multidimensional poverty.
• Multidimensionally, Bangladesh is substantially less poor - but Pakistan
substantially poorer - than would be predicted by income poverty
• Ethiopia is far more multidimensionally poor, and Tanzania much less so,
than predicted by income poverty.
• Most Latin American countries e.g. Brazil rank worse on multidimensional
poverty than on income poverty; but Colombia’s income and MPI poverty
ranks are about same.
Growth and Poverty
• Are rapid growth and poverty reduction in conflict?
• Traditional thinking – in conflict
 Rapid growths bypasses the poor
 Public expenditures slow down growth
 Progressive taxation on the rich (redistribution of income) causes savings to fall.

• Modern thinking – poverty reduction can even accelerate growth (5 reasons)

 Access to credit affects education and entrepreneurship.
 The saving rate among the rich is lower than the middle class, and even the poor.
 Health, nutrition, and education affect economic activity.
 Higher income for the poor raises demand for locally produced goods.
 A reduction of mass poverty is a powerful material and psychological incentive to
widespread public anticipation in the development process.
Economic Characteristics of High-Poverty Groups

• Rural Poverty
Economic Characteristics of High-Poverty Groups
• Rural Poverty
• Women and Poverty
 The earning potential of women is considerably below that of men.
o Women are often paid less for performing similar tasks.
o They are usually barred from high-paying occupations.
o Women are difficult to find jobs in urban areas.
o Rural women have less access to the resources necessary to generate stable incomes.
o Legislation and social custom often prohibit women from owning properties or signing
financial contracts without a husband’s signature.
o Government employment or income-enhancing programs are accessible primarily by men.
 In general, women in female-headed households have less education and lower
incomes.
 Children in female-headed households are less likely to be enrolled in school and
more likely to be working to provide additional income.
 It is important to integrate women in development programs.
Economic Characteristics of High-Poverty Groups
• Rural Poverty
• Women and Poverty
• Ethnic Minorities, Indigenous Populations, and Poverty
Indigenous Poverty in Latin America
Policy Opinions on Income Inequality and Poverty:
Some Basic Considerations

• Altering the Functional Distribution

 lower the relative price of labor and raise the relative price of capital

• Mitigating the Size Distribution

 concentration of asset ownership
 redistribution policies, e.g. land reform

• Moderating (reducing) the Size Distribution at the Upper Levels

 progressive income and wealth taxes
 progressive tax structures often turn out to be regressive taxes

• Moderating (increasing) the Size Distribution at the Lower Levels

 direct transfer payments and the public provision of goods and services
Workfare Program
• Work is required before financial aid is provided.
• Bangladesh Food for Work Program
• India’s Maharashtra Employment Guarantee Scheme
• Workfare represents a better policy than welfare when these criteria met:
 The program does not reduce incentives for the poor to acquire human capital and other assets
 There are greater net benefits of the program’s work output
 It is harder to screen the poor without a workfare requirement

• Poor workers have lower opportunity cost of time (so the economy loses little output when
they work in the program)
• Non-poor workers have higher opportunity cost of time (so they are unlikely to participate
to get the benefits)
• The fraction of the population living in poverty is smaller (so the extra costs of a universal
welfare scheme would be high)
• There is less social stigma of visible workfare participation, so the poor do not suffer
humiliation or be deterred from needed work
(otherwise, a discreet welfare transfer may be preferable)