Inequality
Rising US income inequality is mostly labor income. Every explanation is a shift in a two-panel labor market diagram, low-skill beside high-skill, routed through MRPL = MPL × MR: technology moves MPL, trade moves MR.
The facts
- Three inequalities
- Opportunity, wealth, income. The course does labor income — main driver of rising US income inequality and a source of rising wealth inequality.
- Fractile
- A slice of the distribution defined by percentiles: top 1%, top 10%, top 0.01%.
- Piketty and Saez
- Surveys cannot see the top, so they used IRS tax data back to 1913 with total income and Census household counts. Top shares: high before WWII, low through the 1970s, rising strongly since. The 0.01% rose fastest, driven largely by labor income.
Three explanations, one diagram
| Explanation | Mechanism | High-skill DL | Low-skill DL |
|---|---|---|---|
| Skill-biased tech change | Computers raise MPL of the educated | Out | Barely moves |
| Globalization | US advantage in high-skill goods; their price rises, low-skill-intensive prices fall — MR moves | Out | In |
| Institutions and power | Monopsony; falling private-sector union membership; eroded real federal minimum wage; pay norms | Wage below MRPL | |
Monopsony is ECON 151 material, excluded from Edwards's exams.
The race between education and technology
Skill bias widens the gap; rising educational attainment narrows it. They balanced through the 1950s and 1960s. From around 1970 attainment growth slowed and technology pulled ahead.
Why it matters, and three remedies
- Stakes
- Inequality correlates with poverty and may erode social cohesion and democracy. Case and Deaton: rising mortality and morbidity among less-educated Americans.
- Education
- Not only college. Heckman: returns to human capital investment are highest at the earliest ages for disadvantaged children.
- Help the displaced
- Trade Adjustment Assistance — training, search help, extended benefits, temporary wage subsidies — tiny and little used, Edwards notes. Plus place-based employer subsidies.
- Raise bottom incomes
- Minimum wage, the Earned Income Tax Credit, cash transfers such as a universal basic income.
Rules that earn points
- Draw two labor markets side by side. Nearly every question is one panel shifting.
- Route it through MRPL = MPL × MR. Technology moves MPL; trade moves MR.
- Name the source: Piketty and Saez for shares, Autor for wages, Card for immigration.
- For a remedy, say which curve it shifts.
- Treating inequality as all capital income. The recent US rise is largely labor income.
- Using survey data to talk about the top 1%. It cannot see them.
- Saying the EITC raises the wage firms pay. It lowers the market wage and tops workers up.
Piketty and Saez built their top-income-share series from IRS tax returns rather than household surveys because:
- ATax returns are the only source that goes back to 1913
- BSurveys cannot see the top of the income distribution
- CSurveys systematically overstate incomes at the top
- DOnly tax returns separate labor income from capital income
Answer
B — surveys cannot see the top.
There are too few very rich respondents in a survey sample. Tax records cover them, so the top 1%, 10% and finer fractiles can be backed out from IRS tabulations plus total income and household counts.
A is the trap. The 1913 start date is true and worth writing down, but the reason for switching sources is coverage of the top, not length of the series.
In the globalization explanation of rising inequality, the wage gap widens because trade:
- ARaises MPL for high-skill workers and leaves low-skill MPL alone
- BRaises MR in high-skill-intensive industries and lowers it in low-skill-intensive ones
- CIncreases the supply of low-skill labor in the US
- DErodes the real value of the federal minimum wage
Answer
B — trade works through MR, the output price.
The US exports high-skill-intensive goods and imports low-skill-intensive ones, so the price of the former rises and the price of the latter falls. Through MRPL = MPL × MR, high-skill labor demand shifts out and low-skill labor demand shifts in.
A is the trap: that is skill-biased technological change, which moves MPL. Same divergence, different term in the product.
In a competitive labor market diagram, the Earned Income Tax Credit:
- AShifts labor demand out, raising the wage firms pay
- BShifts labor supply out; the market wage firms pay falls, but workers receive that wage plus the credit
- CShifts labor supply in; the wage rises and employment falls
- DLeaves both curves in place and simply transfers cash
Answer
B — supply shifts out and the market wage falls.
The EITC pays people to work, so labor supply shifts from S₁ to S₂. Employment rises to L₂, the market wage falls to W₂, and workers take home W₂ + EITC, which is above the original W₁. Because the market wage falls, employers capture part of the credit.
A is the trap. The credit goes to workers, not to firms buying labor, so it is a supply-side shift. Saying demand rises gets the incidence exactly backwards.
"The race between education and technology" explains rising US inequality after about 1970 by:
- ATechnology slowing down while educational attainment kept rising
- BGrowth in educational attainment slowing while skill-biased technological change continued
- CBoth forces accelerating together
- DThe collapse of private-sector union membership alone
Answer
B — education slowed, technology did not.
Demand-side skill bias widens the gap; rising supply of educated workers narrows it. The two roughly balanced through the 1950s and 1960s. After about 1970 attainment growth slowed and the demand-side force won.
A is the direction flip. If technology had slowed while education kept rising, the gap would have narrowed.
Use labor market diagrams to show how skill-biased technological change raises wage inequality. Then state what real-wage evidence supports it.
Model answer
Two panels, each with its own downward-sloping labor demand and upward-sloping labor supply.
Computers and related technology raise MPL for highly educated workers, so in the high-skill panel demand shifts out from D₁ to D₂: WH and LH both rise. In the low-skill panel demand barely moves, so WL is roughly unchanged. The gap WH − WL widens.
The evidence: Autor's series on real wages of full-time male workers by education level shows exactly this divergence.
- Two separate markets, not one. A single diagram cannot show a gap.
- Shifting demand, not supply, in the high-skill panel.
- Saying the low-skill panel is roughly unchanged rather than shifting it inward.
- Naming MPL as the channel — technology raises the marginal product of educated workers.
- Both equilibria labelled; the widening gap stated explicitly.
Why did Piketty and Saez use tax returns instead of surveys, and what did they find about US top income shares over the twentieth century?
Model answer
Surveys cannot see the top of the income distribution — too few very rich respondents, and underreporting. US income tax data run back to 1913, so Piketty and Saez combined IRS tabulations of taxpayers by income bracket with total income and Census household counts to compute the share going to the top 1%, top 10% and finer fractiles.
The finding: top shares were high before World War II, low through the 1970s, and have risen strongly since, with the 0.01% rising fastest. The recent rise is driven substantially by labor income, which is why the labor market diagram is the right tool.
- Naming the survey limitation as coverage of the top, not sample size in general.
- The three-part shape — high, low, rising — in that order.
- Saying the recent rise is labor income, not capital income.
- Using the word fractile or otherwise showing you know top shares are percentile slices.