🔒 Private Site

This site is password-protected.

Chapter 15: Poverty and Economic Inequality

Why do millions of Americans live below the poverty line despite decades of antipoverty programs? How do we measure inequality, and what can government do about it? This chapter explores the measurement of poverty, the perverse incentives within welfare systems (the “poverty trap”), the U.S. safety net, methods for measuring income inequality (quintiles and Lorenz curves), and the tradeoffs inherent in policies that seek greater equality.


1. Drawing the Poverty Line

Poverty is measured by the number of people who fall below a certain level of income — the poverty line — that defines the income needed for a basic standard of living. Income inequality compares the share of total income (or wealth) that different groups receive.

1.1 The Orshansky Method

The U.S. poverty line traces back to Mollie Orshansky, who in 1963 published “Children of the Poor” while working at the Social Security Administration. Her approach:

  1. Calculate the cost of a nutritionally adequate diet for a given family size
  2. Observe that the average family spent one-third of its income on food
  3. Multiply the food budget by 3 to get the poverty line

The government adjusts the dollar amounts over time to maintain purchasing power.

Family Size 2021 Poverty Line
1 person $12,880
4 people $26,500

1.2 U.S. Poverty Rates Over Time

  • The poverty rate declined through the 1960s
  • Rose in the early 1980s and early 1990s
  • Generally slightly lower since the mid-1990s
  • In no year has the rate fallen below 10.5% — roughly 1 in 9 Americans
  • Peaked at 15.1% in 2010; dropped to 10.5% in 2019; rose to 11.4% in 2020 (COVID-19)

1.3 Poverty Rates by Group (2020)

Group Poverty Rate
Females 12.6%
Males 10.2%
White (Non-Hispanic) 8.2%
Black 19.5%
Hispanic 17.0%
Under age 18 16.1%
Ages 18–64 10.4%
Ages 65+ 9.0%

Important nuance: While Hispanic and African American populations have higher poverty rates, the largest absolute number of people below the poverty line are White — because White people are a larger share of the total population.

1.4 Global Poverty Lines

The World Bank sets two poverty lines for low-income countries:

  • $1.90/day per person
  • $3.20/day per person

For comparison, the U.S. 2015 poverty line works out to about $18.35/person/day.

Country Year < $1.90/day < $3.20/day
Brazil 2019 4.6% 9.1%
China 2017 0.5% 5.4%
Egypt 2017 3.8% 28.9%
India 2011 22.5% 61.7%
Mexico 2018 1.7% 6.5%
Nigeria 2018 39.1% 71.0%

1.5 Criticisms of the Poverty Line

  • Should there be a national line when costs vary widely? (Median household income: $109,113 in New Jersey vs. $59,701 in Mississippi, 2017)
  • Based on cash income only — does not account for non-cash benefits (Medicaid, food aid, housing assistance)
  • Government statisticians at the Census Bureau continuously research these questions

2. The Poverty Trap

Poverty trap: An antipoverty program structured so that government benefits decline substantially as people earn more income — as a result, working provides little or no financial gain.

2.1 How the Trap Works

Example — Single mother, two children, $8/hour:

Without government assistance, working 2,000 hours/year (40 hrs/wk × 50 wks) earns $16,000.

Suppose a program guarantees $18,000 but reduces support $1 for every $1 earned:

Hours Worked Earnings Gov’t Support Total Income
0 $0 $18,000 $18,000
500 $4,000 $14,000 $18,000
1,000 $8,000 $10,000 $18,000
1,500 $12,000 $6,000 $18,000
2,000 $16,000 $2,000 $18,000
2,500 $20,000 $0 $20,000

Working 2,000 hours produces the same income as working 0 hours. Only at 2,300+ hours does income exceed $18,000 — and even then by just $400.

2.2 Additional Factors Worsening the Trap

  • Working mothers face extra expenses: clothing, transportation, child care
  • Non-working individuals fail to build job experience and contacts, making future employment less likely
Hours Worked per Year Total Income ($) 0 500 1,000 1,500 2,000 $0 $10K $18K $26K No assistance $1-for-$1 phase-out (trap!) P 50¢ phase-out S Poverty Trap: Budget Constraints Red = no incentive to work | Green = work pays

2.3 Loosening the Poverty Trap

Reduce the phase-out rate. Instead of $1-for-$1, reduce benefits by 50 cents for every $1 earned:

Hours Worked Earnings Gov’t Support Total Income
0 $0 $18,000 $18,000
500 $4,000 $16,000 $20,000
1,000 $8,000 $14,000 $22,000
1,500 $12,000 $12,000 $24,000
2,000 $16,000 $10,000 $26,000
2,500 $20,000 $8,000 $28,000

Now every hour of work brings an effective wage of $4/hour (half of $8). The vertical intercept rises to $28,000.

Trade-off: A slower phase-out preserves work incentives but costs more, since benefits extend to higher income levels. Some recipients may work fewer hours if the higher budget line still gives them more income than before (point S vs. point P on the budget constraint).

Additional mechanisms to reduce the trap:

  • Work requirements as a condition for benefits
  • Time limits on benefits

3. The Safety Net

Safety net: The group of government programs that provide assistance to people at or near the poverty line, offering protection for those without jobs or income.

3.1 Temporary Assistance for Needy Families (TANF)

Feature AFDC (pre-1996) TANF (1996–present)
Enacted by Social Security Act (1935) Personal Responsibility and Work Opportunity Reconciliation Act (1996)
Funding Open-ended federal matching Fixed block grants to states
Work requirements None Required — most recipients must work or attend school
Time limits None 5-year lifetime limit on federal benefits
State flexibility States set benefit levels States can use funds for any antipoverty program

Key outcomes:

  • Families receiving payments fell from 4.8 million (1995) to 1.0 million (2020) — a decline of nearly 80%
  • Benefits vary widely: $862/month in New Hampshire vs. $146/month in Mississippi (2020, single mother with one child)
  • Total spending peaked at ~$35 billion (2010), fell to ~$30 billion (2020)

3.2 Earned Income Tax Credit (EITC)

EITC (1975): A method of assisting the working poor through the tax system. Benefits increase with earned income up to a point, then phase out slowly.

  • In 2021: ~25 million eligible workers/families received about $60 billion
  • Credit ranges from $1,502 to $6,728 depending on filing status, income, and number of children
  • Average EITC: ~$2,411 per family with children
  • Phase-out example (single parent, 2 children, 2013): credit stable from $13,430–$17,530, then reduced by 21.06 cents per $1 earned, fully phased out at $46,227
  • Popular because it rewards work rather than discouraging it

3.3 Supplemental Nutrition Assistance Program (SNAP)

  • Federally funded since 1964 (formerly “food stamps”)
  • Recipients receive a debit-like card to purchase food
  • Households expected to spend ~30% of net income on food; SNAP covers the gap
  • In 2021: about 41.5 million people received benefits; total ~$108 billion (~$287/person/month)
  • Participation surged 70% between 2007 and 2011 (Great Recession); rose another 5.8M from 2019–2021 (COVID-19)
  • Has its own work requirements and time limits

Economics of SNAP cards vs. cash: Even though SNAP cards can only buy food, families may use them to free up cash previously spent on food — effectively spending the aid on other goods. Economists consider SNAP functionally similar to a cash transfer.

3.4 Medicaid

  • Created in 1965 as a joint federal-state program
  • Provides medical insurance for certain low-income individuals (families with children, elderly, disabled)
  • As of 2014: ~69.7 million enrolled
    • Children: 47% of enrollees, 21% of spending
    • Seniors: 9% of enrollees, 20% of spending
    • Blind/disabled: 16% of enrollees, 44% of spending
  • Poverty-trap risk: Low-paying jobs may disqualify a family from Medicaid without offering employer health insurance
  • Many states expanded eligibility to 135%–185% of the poverty line in the 1980s and 1990s

3.5 Other Safety Net Programs

Program Purpose
WIC Food assistance for pregnant women, infants, and children
School lunch/breakfast Subsidized meals for children from low-income families
Low Income Home Energy Assistance Help with home heating bills
Housing assistance Subsidized rent
Supplemental Security Income (SSI) Cash support for disabled and elderly poor

4. Income Inequality: Measurement and Causes

4.1 Quintile Measurement

Quintile: Dividing a ranked population into five groups of equal size (each 20%) and comparing the share of total income each group receives.

U.S. Income Distribution by Quintile, 1967–2020:

Year Bottom 20% Second 20% Third 20% Fourth 20% Top 20% Top 5%
1967 4.0% 10.8% 17.3% 24.2% 43.6% 17.2%
1980 4.2% 10.2% 16.8% 24.7% 44.1% 16.5%
1990 3.8% 9.6% 15.9% 24.0% 46.6% 18.5%
2000 3.6% 8.9% 14.8% 23.0% 49.8% 22.1%
2010 3.3% 8.5% 14.6% 23.4% 50.3% 21.3%
2020 3.0% 8.1% 14.0% 22.6% 52.2% 23.0%

Key trend: From 1980 to 2020, the top 20%’s share rose from 44.1% → 52.2% (+8.1 pp), while the bottom 20%’s share fell from 4.2% → 3.0% (–1.2 pp). The top 5% alone gained from 16.5% → 23.0%.

4.2 The Lorenz Curve

Lorenz curve: A graph comparing the cumulative share of income actually received (vertical axis) against the cumulative share of population (horizontal axis). A 45° line represents perfect equality.

Calculating Lorenz Curve Points (2020 data):

Cumulative Population Cumulative Income (1980) Cumulative Income (2020)
20% 4.2% 3.0%
40% 14.4% 11.1%
60% 31.2% 25.1%
80% 55.9% 47.7%
100% 100.0% 100.0%
  • A more unequal distribution loops farther from the 45° line
  • The 2020 Lorenz curve is farther from the line than 1980 → inequality increased
Cumulative % of Population Cumulative % of Income 0 20% 40% 60% 80% 100% 0 25% 50% 75% 100% Perfect Equality 1980 2020 A B Growing inequality U.S. Lorenz Curves: 1980 vs. 2020 Gini Coefficient = A / (A + B)

4.2.1 The Gini Coefficient

Gini coefficient — a single number that summarizes inequality from the Lorenz curve:

\[G = \frac{A}{A + B}\]

where A = area between the 45° line and the Lorenz curve, and B = area below the Lorenz curve.

  • $G = 0$ → perfect equality (everyone earns the same)
  • $G = 1$ → perfect inequality (one person earns everything)

International Gini Coefficients (World Bank):

Country Gini Interpretation
Sweden 0.28 Low inequality
Germany 0.32 Low–moderate
Canada 0.33 Moderate
India 0.36 Moderate
United States 0.39 High for a developed nation
China 0.38 High
Mexico 0.45 High
Brazil 0.49 Very high
South Africa 0.63 Extreme

Worked Example — Approximate Gini from Quintile Data

Using the 2020 U.S. quintile data and the trapezoid method:

Cumulative income shares: 0, 3.0, 11.1, 25.1, 47.7, 100 (as fractions of 100).

Area under Lorenz curve (B) using trapezoids (each width = 0.2):

\[B = 0.2 \times \left[\frac{0 + 0.030}{2} + \frac{0.030 + 0.111}{2} + \frac{0.111 + 0.251}{2} + \frac{0.251 + 0.477}{2} + \frac{0.477 + 1.0}{2}\right]\] \[B = 0.2 \times [0.015 + 0.0705 + 0.181 + 0.364 + 0.7385] = 0.2 \times 1.369 = 0.2738\]

Area under 45° line (A + B) = 0.5.

\[G = \frac{A}{A+B} = \frac{0.5 - 0.2738}{0.5} = \frac{0.2262}{0.5} = \mathbf{0.452}\]

This approximation is higher than the official figure (0.39) because quintile data loses granularity within groups — finer data produces a smoother curve closer to the 45° line.

4.3 International Comparison

Country Year Bottom 20% Second 20% Third 20% Fourth 20% Top 20%
United States 2020 3.0% 8.1% 14.0% 22.6% 52.2%
Germany 2016 7.6% 12.8% 17.1% 22.8% 39.6%
Brazil 2019 3.1% 7.4% 12.3% 19.4% 57.8%
China 2016 6.5% 10.7% 15.3% 22.2% 45.3%
India 2011 8.1% 11.7% 15.2% 20.5% 44.4%
Nigeria 2018 7.1% 11.6% 16.2% 22.7% 42.4%

The U.S. has greater inequality than Germany and many comparable high-income nations, and even more than some low/middle-income countries like India and Nigeria. Latin America (Brazil, Mexico) has the highest inequality.

4.4 Causes of Growing Inequality

Cause 1: Changing Household Composition (~50% of the increase)

  • Married women in the labor force: 41% (1970) → 58.6% (2019)
  • Assortative mating: doctors marry doctors, executives marry executives → more two-high-earner households
  • Single-parent families: ~23% of all U.S. families in 2021 → higher poverty rates
  • These shifts push the top and bottom of the distribution further apart

Cause 2: Shift in Wage Distribution — “Winner-Take-All” Labor Markets

  • College-educated workers (ages 25–34) earned 1.59× as much as high school graduates in 1995 → 1.85× in 2020
  • Supply of college graduates increased: 840,000 bachelor’s degrees (1970) → 2.0 million (2018–2019), up 138%
  • Despite this supply increase, wages for skilled workers rose → demand must have shifted right even faster
  • Drivers of rising demand for skilled labor:
    • Information and communications technology — benefits high-skilled workers disproportionately
    • Globalization — opens markets for high-skill services worldwide while exposing low-skill workers to global competition
    • “Star” premium — global demand for top CEOs, athletes, actors pushes salaries far above productivity differences

The “race” metaphor: We can view the market for high-skilled labor as a race between supply forces (more education) and demand forces (technology, globalization). When demand grows faster than supply, inequality widens.


5. Government Policies to Reduce Income Inequality

5.1 Why Some Inequality Is Expected

  • Life-cycle effects: Low earnings in early jobs → higher in middle age → lower in retirement
  • Preference differences: Some choose high-effort/high-income careers; others value leisure
  • Income vs. wealth: Wealth (sum of all assets minus debts) is even more unequally distributed than income

Income is a flow of money received (monthly/annually). Wealth is the stock — total value of all assets minus debts. The Federal Reserve’s Survey of Consumer Finance (published every 3 years) tracks wealth distribution.

5.2 Three Main Policy Tools

(A) Redistribution

  • Fund programs (TANF, EITC, SNAP, Medicaid) through a progressive tax system — the rich pay a higher percentage
  • 2018 data: Top 1% had average income of $1,679,000 and paid an average federal tax rate of 25.4% (effective rate: 20.4%)
  • Bottom two quintiles paid negative effective income taxes (thanks to provisions like the EITC)

(B) Ladder of Opportunity

Children College Level Adults
Improved day care Widespread loans and grants Retraining / new skills
Preschool enrichment Community colleges → research universities Anti-discrimination laws
Improved public schools
After-school/community activities
Internships & apprenticeships

(C) Estate (Inheritance) Taxes

  • The U.S. estate tax in 2022 applied only to inheritances exceeding $12.06 million — affecting a tiny percentage of the population
  • Debate: Should you be free to pass wealth to heirs? vs. Should society limit intergenerational wealth concentration?

5.3 The Tradeoff Between Incentives and Equality

Two views of the tradeoff:

(a) Pessimistic view: Any move toward equality (A → B) reduces economic output.

(b) Optimistic view:

  • C → D: Some policies (e.g., public education) can increase both output and equality
  • D → E: A range where countries with similar output levels (U.S., Canada, EU, Japan) can choose different inequality levels with little output impact
  • E → F: Only extreme redistribution reduces output through diminished incentives

Key insight: Moderate equality-promoting policies may sustain political support for market economies. Without them, citizens might demand rigid labor laws, price controls, or trade restrictions that would reduce output more severely.


6. Key Takeaways

  1. The U.S. poverty line originated in 1963 (Mollie Orshansky) as 3× the food budget; it has never fallen below 10.5% of the population
  2. The poverty trap occurs when $1 earned means ≈$1 lost in benefits — it can be loosened by slower phase-outs, work requirements, and time limits
  3. The safety net includes TANF (work-based, time-limited), EITC (rewards work), SNAP ($108B, 41.5M), Medicaid (69.7M), and other programs
  4. Income inequality is measured by quintiles (top 20% received 52.2% of income in 2020) and Lorenz curves (farther from 45° = more inequality)
  5. Growing inequality results from changing household composition (~50%) and rising skill premiums driven by technology and globalization
  6. Policy tools — redistribution, ladder of opportunity, estate taxes — involve a tradeoff between incentives and equality, though moderate policies may improve both

7. Practice Questions

Q1. What was Mollie Orshansky’s method for determining the poverty line?

Answer She calculated the cost of a nutritionally adequate diet for a given family size and multiplied it by 3, since the average family spent one-third of its income on food.

Q2. A single parent earning $10/hour can work up to 2,000 hours/year. The government guarantees $15,000 but reduces support $1-for-$1 with earnings. At what point does working yield more than $15,000?

Answer The parent earns $20,000 at full-time work (2,000 × $10). Since benefits are reduced $1 for $1, total income is $15,000 until earnings exceed $15,000 — i.e., at 1,500+ hours. At 1,500 hours: $15,000 earned + $0 support = $15,000. So the person must work more than 1,500 hours to exceed the guarantee. This is the poverty trap in action.

Q3. How did TANF (1996) differ from the old AFDC program it replaced?

Answer TANF imposed work requirements (most recipients must work or attend school), set a 5-year lifetime limit on federal benefits, changed federal funding from open-ended matching to fixed block grants, and gave states wider flexibility to use funds for any antipoverty objective.

Q4. Why is the EITC popular with both economists and the general public?

Answer Because it increases the payment received for work (rewards labor), phases out slowly (minimizing the poverty trap), and provides the largest benefits to those who are working — aligning incentives with policy goals.

Q5. Explain why giving SNAP cards instead of cash may not actually change what families spend money on.

Answer If a family was already spending $2,500/year on food and receives $1,000 in SNAP, it can continue spending $2,500 on food (using SNAP for part of it) and redirect the $1,000 in freed-up cash to other goods. SNAP effectively becomes a cash transfer.

Q6. In 2020, the top 20% of U.S. households received 52.2% of total income. What share did the bottom 40% receive?

Answer Bottom 20% (3.0%) + Second 20% (8.1%) = bottom 40% received **11.1%** of total income — about one-fifth of the top quintile's share alone.

Q7. What is the difference between the Lorenz curve and the 45° line of perfect equality?

Answer The 45° line represents perfect equality — e.g., the bottom 20% receives exactly 20% of income. The Lorenz curve plots the actual cumulative income distribution. The farther the Lorenz curve bows below the 45° line, the greater the inequality.

Q8. How can a country experience greater income equality yet still have high poverty rates?

Answer If everyone's income declines proportionally, inequality stays the same or even decreases, but more people fall below the poverty line. Equality measures distribution, not absolute levels.

Q9. Explain why the college wage premium increased (1.59× to 1.85×) despite a large increase in the supply of college graduates.

Answer The demand for high-skilled labor shifted right even faster than supply — driven by information technology, globalization, and "winner-take-all" markets. The combination of both shifts resulted in higher wages for skilled workers despite more graduates.

Q10. A group of 10 people earn: $10K, $12K, $16K, $18K, $24K, $24K, $36K, $50K, $80K, $100K. Calculate the income share of the bottom and top quintiles.

Answer Total = $370K. Bottom quintile (lowest 2 earners): $10K + $12K = $22K → 22/370 = **5.9%**. Top quintile (highest 2 earners): $80K + $100K = $180K → 180/370 = **48.6%**.

Q11. What are the two main explanations economists give for rising U.S. income inequality since the late 1970s?

Answer (1) **Changing household composition** — more two-high-earner couples at the top and more single-parent households at the bottom (accounts for ~50% of the increase). (2) **Rising demand for high-skilled labor** relative to low-skilled, driven by technology and globalization ("winner-take-all" markets).

Q12. Explain the “optimistic” view of the equality–output tradeoff. Why might moderate redistribution actually increase economic output?

Answer Programs like public education redistribute income (funded by taxes, benefiting low-income families disproportionately) while also building human capital that drives growth. Additionally, moderate redistribution sustains political support for market economies, preventing citizens from demanding rigid regulations that would reduce output more severely.

Q13. Using the following quintile data, calculate the Gini coefficient using the trapezoid approximation.

Quintile Income Share
Bottom 20% 5%
Second 20% 10%
Third 20% 15%
Fourth 20% 25%
Top 20% 45%
Answer Cumulative shares: 0, 5, 15, 30, 55, 100 (as percentages). Area under Lorenz (B) = $0.2 \times [\frac{0+0.05}{2} + \frac{0.05+0.15}{2} + \frac{0.15+0.30}{2} + \frac{0.30+0.55}{2} + \frac{0.55+1.0}{2}]$ $= 0.2 \times [0.025 + 0.10 + 0.225 + 0.425 + 0.775] = 0.2 \times 1.55 = 0.31$ $G = \frac{0.5 - 0.31}{0.5} = \frac{0.19}{0.5} = \mathbf{0.38}$ This is comparable to the U.S. Gini coefficient, indicating relatively high inequality.

Q14. A government program guarantees $20,000 for a family of four. Under Plan A, benefits are reduced $0.80 per $1 earned. Under Plan B, benefits are reduced $0.40 per $1 earned. For a parent earning $12/hr working 1,500 hours: (a) Calculate total income under each plan. (b) Which plan better preserves work incentives? (c) Which plan costs the government more?

Answer Earnings = $12 × 1,500 = $18,000. **(a)** Plan A: Benefit reduction = $18,000 × 0.80 = $14,400. Remaining benefit = $20,000 − $14,400 = $5,600. Total = $18,000 + $5,600 = **$23,600**. Plan B: Benefit reduction = $18,000 × 0.40 = $7,200. Remaining benefit = $20,000 − $7,200 = $12,800. Total = $18,000 + $12,800 = **$30,800**. **(b)** Plan B preserves work incentives better — each additional dollar earned loses only 40¢ in benefits (effective wage = $7.20/hr) vs. Plan A where each dollar loses 80¢ (effective wage = $2.40/hr). **(c)** Plan B costs the government more ($12,800 vs. $5,600 per family) and extends benefits to higher income levels. Plan B benefits don't phase out until earnings reach $20,000/0.40 = $50,000, vs. Plan A at $20,000/0.80 = $25,000.

Q15. Country X has a Gini coefficient of 0.25 and Country Y has a Gini coefficient of 0.55. (a) Which country has greater income inequality? (b) Can you conclude which country has higher poverty rates? (c) If Country Y implements a progressive tax that moves its Gini to 0.40, describe the change on a Lorenz curve diagram.

Answer **(a)** Country Y (Gini 0.55) has much greater income inequality. **(b)** No. Gini measures *relative* distribution, not absolute income. Country Y could be a wealthy nation (like the U.S.) where even low-income earners are above the poverty line, while Country X could be a poor nation where incomes are equally distributed but all very low. **(c)** Country Y's Lorenz curve would shift **inward** (closer to the 45° line). The area A between the 45° line and the Lorenz curve shrinks. Specifically, the curve moves from enclosing 55% of the triangle (G=0.55) to 40% (G=0.40), representing a meaningful reduction in inequality.

8. Glossary

Term Definition
Poverty Being below the income level needed for a basic standard of living
Poverty line The specific income threshold for a basic standard of living
Poverty rate Percentage of the population living below the poverty line
Poverty trap When government benefits decline so steeply with earned income that working provides little financial gain
Safety net Government programs that assist people at or near the poverty line
TANF Temporary Assistance for Needy Families — replaced AFDC in 1996, with work requirements and time limits
EITC Earned Income Tax Credit — assists the working poor through tax refunds that phase out slowly
SNAP Supplemental Nutrition Assistance Program (“food stamps”) — debit cards for food purchases
Medicaid Joint federal–state health insurance for certain low-income individuals (1965)
Quintile Dividing a population into fifths to compare income distribution shares
Lorenz curve Graph comparing cumulative income received to a perfectly equal distribution (45° line)
Progressive tax system Tax system where the rich pay a higher percentage of income in taxes
Redistribution Taking income from higher-income groups and providing it to lower-income groups
Effective income tax Total taxes paid ÷ total income
Estate tax Tax imposed on the value of an inheritance
Income A flow of money received, typically measured monthly or annually
Wealth Sum of all asset values minus all debts
Income inequality When different groups receive disproportionate shares of total income or wealth

← Back to Economics & Finance Hub