Topic 2 - Poverty and Inequality Flashcards
Extreme poverty (for low income countries) definition, according to the World Bank
% of people living with under $1.90 a day in 2022 PPP dollars
This number is based on the cheapest way to achieve min caloric requirements (~2100 calories per day) and other locally relevant minimum basic needs.
Line is updated in accordance with nominal changes in value of the dollar.
Head Count Index (HCI)
Proportion of individuals living under the poverty line
HCI = H = Q/N
where
- Q = number of people under the absolute poverty line z
- N = population
- H = measure of extent of absolute poverty
Limitations of head count index
- need to agree on a poverty line
- looks only at extent of poverty (number of people below poverty line) but depth of poverty (how far the incomes of these people are below the poverty line)
Poverty gap (world bank definiton)
Measures the extent to which individuals fall below the poverty line (the poverty gaps) as a proportion of the poverty line. The sum of these poverty gaps gives the minimum cost of eliminating poverty, if transfers were perfectly targetted
Poverty gap (PG) equation
PG = [Q(z-yq)]/z
where
z = absolute poverty line
yq = mean income of the poor
Q = number of people below the absolute poverty line (z)
Why care about poverty?
- Equity
- care about human welfare
- social costs associated w poverty - Future potential
- living standards + economy improve if people able to reach fill potential - Distributive justice
- maximize equality of resources
- veil of ignorance (maximin) - want standard of living to be as high as possible - Limits efficiency
- reduced productivity = bad for growth and development
- low savings = reduced economic growth and economic instability
- can lead to market failures
- risk and time preferences (will not invest in future if cant meet current needs) - political stability
- poverty –> unstable politics –> less productivity –> bad for economy
Why do we normalize (divide by z) the poverty gap equation?
the equation becomes
- unitless
- comparable across countries, currencies
Average poverty gap
APG = [Q(z-yq)]/[nz]
–> APG = H[(z-yq)/z)]
Pros and cons of measuring poverty using income vs consumption
Income pros
1. in an ideal world, can be measured precisely
Income cons:
1. measurement errors lead to income being under-reported
–> disincentives to reveal the truth (lower income –> lower taxes)
–> reporting bias and recall bias
–> in informal economy where large sector is not monitored or is self employment, measuring income can be very difficult
–> survey data may be inaccurate. Difficult to reach ultra rich and ultra poor.
- requires detailed survey data
Consumption pros
1. less measurement error
- some measurement error because associated with peoples memories but less bad than for income measurements
Consumption cons
1. believed to be a better indicator of permanent income
2. requires detailed survey data
in general - consumption data is more accurate
Multidimensional poverty index (OPHI)
A person is poor if they are deprived in 1/3 or more of weighted indicators
10 indicators categorized into three dimensions of poverty (health, education, living standard)
10 indicators:
Health
1. nutrition
2. child mortality
Education
3. years of schooling
4. school attendance
Living Standard
5. cooking fuel
6. improved sanitation
7. safe drinking water
8. electricity
9. flooding
10. Assets (ex livestock)
Monetary poverty
- reduced consumption –> higher poverty
- often positively related to multidimensional poverty
Income distribution
The amounts of income received by the rich, poor, and middle class individuals or families
- often interpreted as direct measure of welfare
- used to measure inequality
- can be expressed in terms of income deciles (tenths), quintiles (fifths), or quartiles (fourths)
Social welfare
How well off a society is without considering how evenly income is distributed among its citizens
the lorenz curve
Depicts the cumulative distribution of income
y axis = percent of income
x axis = percent of population
in an egalitarian society, income distribution creates a diagonal line at 45 degrees
In real societies, income distribution is represented by the lorenz curve, which is a bowed line located beneath this 45 degree line
The greater the curvature of the lorenz curve, the greater the degree of inequality .
constructing lorenz curves
calculate the cumulative shares of income for each decile (0%, 10%, …, 100%)
- add % share of total income in previous decile to % share in current decile to get cumulative share of total income in current decile
Plot cumulative shares for each decile of population to create lorenz curve
Gini coefficient
Gini (G) is a way to collapse information from the lorenz curve into a single number. This number gives income distribution but does not allow for comparison of incomes between countries.
G = twice the area between the diagonal line and the lorenz curve
G=0: perfect equality
G=1: perfect inequality
In reality, G=1-2S
where S = area under lorenz curve
comparing lorenz curve and gini coefficient
If gini coefficient is the same for different income distributions, the lorenz curves cross. If these lines cross, you cannot compare inequality.
Evolution of distribution of income and world being among world citizens 1820-1992
Study by Bourguignon and Morrisson found
- worsening of world income distribution from 18201-wwII
- stabilized or worsened more slowly since wwII
- 19th century - inequality due to within-country inequality
- 20th century - inequality due to difference between countries
- currently seeing trend reversal since article published
Kuznets curve
U-shaped relationship between development/ growth and inequality
possible explanation: shifting out of agriculture (but little empiracle evidence for this)
Poverty targeting
Designing policies specifically aimed at reducing poverty among beneficiaries
Why target policies toward the poor?
- maximize coverage
- Minimize leakage to the non-poor
tradeoffs in anti-poverty policies
- equity vs efficiency
- costs more to reach higher accuracy in reaching the poor - administrative costs vs asymmetric information
- administrative costs: government/agency and costs for the poor of filling out forms, traveling etc
- asymmetric information:
–> adverse selection: attracting the non-poor
–> moral hazard: program transfer may alter behaviour
Mechanisms for targeting
- Indicators: based on some characteristics (eg income, land, age, gender, etc)
- means testing: if a persons income is below a certain level per year the person will receive benefits
- strong incentive effects: people will identify as having certain characteristics to receive benefits - self-targetting: self-identity (eg workfare, placement of benefits office in urban informal settlements, food subsidies, etc)
- participatory wealth rankings: community based
- used especially in rural areas where difficult to get individual income based measures
Perfect targeting anti-poverty policy
Want to give everyone below poverty line exactly enough to bring up to poverty line. Financed through taxation of non-poor
transfer function
T=f(y) where y is individual income, given a poverty line z
Total Transfer = sum from i=1 to Q (z-yi)
where
z = absolute poverty line
yi = mean income of individual
Q = number of people below the absolute poverty line (z)
pros
- most equitable
cons
- administratively costly
- incentive effects (people working less hard get same income –> disincentivizes hard work)
- need a good tax system
- political costs
