Transparency & Rigor

Data Methodology & Sources

Every data series on this platform has known limitations, series breaks, and contested interpretations. This page documents them fully. Scholarly honesty about measurement uncertainty is a prerequisite for sound legal and policy analysis.

Overview

Three Primary Data Series

This platform draws on three primary data series, each measuring a distinct dimension of economic distribution: (1) the Piketty-Saez top income share series, measuring the concentration of fiscal income among the top decile and percentile since 1917; (2) the Census Bureau Gini coefficient series, measuring overall household income inequality since 1967; and (3) the official poverty rate, measuring the share of the population below the Orshansky poverty thresholds since 1959. A fourth series — unofficial pre-1959 poverty estimates — is available in the Data Explorer with prominent warnings and should not be used for quantitative analysis without careful reading of the caveats below.

Note on causation: This platform presents correlations between legislative events and distributional outcomes. Establishing causation requires quasi-experimental methods, instrumental variables, or natural experiments. The charts and timelines here are designed to prompt inquiry, not to establish causal claims. Readers are directed to the empirical literature cited in the Further Reading section for rigorous causal analysis.

Series 1

Piketty-Saez Top Income Shares

Coverage: 1917–2022  |  Updated: March 2024  |  Unit: Tax units  |  Frequency: Annual
Primary Citation
Thomas Piketty & Emmanuel Saez, “Income Inequality in the United States, 1913–1998,” Quarterly Journal of Economics 118:1 (February 2003), pp. 1–41. Updated series available at gabriel-zucman.eu (March 2024 update).
What Is Measured
Pre-tax, pre-transfer fiscal income of tax units, as reported to the IRS via adjusted gross income (AGI) plus certain adjustments (retirement income, employer-paid benefits partial). The series measures the share of total fiscal income accruing to the top 10% and top 1% of tax units ranked by income.
What Is Not Measured
The series does not include: Social Security benefits (partially), Supplemental Nutrition Assistance (SNAP), Medicaid and Medicare benefits, housing vouchers, the Earned Income Tax Credit (EITC), employer-paid health insurance (except partial), imputed rent, or undistributed corporate profits (except in the national income versions). It measures market income before redistribution.
Income-Timing Artifacts
Two well-documented spikes are measurement artifacts rather than genuine changes in inequality. 1986: The Tax Reform Act reduced the top marginal rate from 50% to 28% effective January 1, 1987. High-income taxpayers accelerated income realization into calendar year 1986, producing an artificial spike visible in both the top 10% and top 1% series. 2012: Anticipating the expiration of the Bush-era top rates in 2013, high-income taxpayers again accelerated income realization, producing a 2012 spike (top 1% reached 23.7%). These artifacts should be interpreted as income timing, not genuine changes in the underlying distribution.
The Auten-Splinter Critique
Gerald Auten (U.S. Treasury) and David Splinter (Joint Committee on Taxation), “Income Inequality in the United States: Using Tax Data to Measure Long-Term Trends,” Journal of Political Economy 132:7 (July 2024), pp. 2179–2227. Auten and Splinter adjust the Piketty-Saez series for: taxes paid, government transfers received, the growing number of non-filers, and the attribution of corporate retained earnings and government spending. They find that post-tax-and-transfer top income shares have increased far less than the Piketty-Saez pre-tax series suggests. The debate is ongoing; this platform presents the Piketty-Saez series as the most widely cited and longest internally consistent series, while acknowledging that the after-tax picture is substantially different.
Known Limitations
  • Relies on IRS administrative data, which reflects legal tax avoidance strategies and changes in tax law that affect what is reported as income
  • Unit of analysis is the tax unit, not the individual or household; filing patterns have changed over time
  • Does not capture non-filers (disproportionately low-income), potentially understating bottom income shares
  • Capital gains treatment varies between series (excl. and incl. versions); this platform uses excl. for top 10% and incl. for top 1%
  • National income and product account (NIPA) adjustments in newer versions create series breaks from older versions

Series 2

Census Bureau Gini Coefficient

Coverage: 1967–2024  |  Updated: September 2025  |  Unit: Households  |  Frequency: Annual
Primary Citation
U.S. Census Bureau, Current Population Survey Annual Social and Economic Supplement (CPS ASEC), Historical Income Tables, Table H-4 (Gini Ratios for Households, by Race and Hispanic Origin: 1967 to Present). Also available via FRED as series GINIALLRH.
What Is Measured
The Gini coefficient is a summary measure of inequality ranging from 0 (perfect equality, all households have equal income) to 1 (maximum inequality, all income held by one household). The Census series measures money income of households, defined as pre-tax cash income including wages, salaries, self-employment income, Social Security and other transfers, interest, dividends, and rental income. It excludes capital gains and non-cash benefits.
Series Breaks — Use With Caution
The Census Gini series contains three documented discontinuities that affect comparisons across time:
  • 1993: Introduction of computer-assisted telephone interviewing (CATI) and expansion of the CPS sample. Increased recall and reporting, causing a measured upward shift in income and inequality that is partly or entirely a measurement artifact. The Census Bureau estimates the redesign raised the measured Gini by approximately 0.004–0.006.
  • 2013–14: Redesign of the CPS ASEC income questions, adding new income categories and improving measurement of investment income. Caused a measurable upward shift in reported inequality. The Census Bureau published parallel estimates on both old and new processing for 2013 to document the break.
  • 2017–18: Processing updates to how certain income items were coded changed measured Gini estimates. Less well-documented than the prior breaks.
Top-Coding
The CPS ASEC top-codes certain high income values (e.g., wage and salary income above $1.1 million is topcoded). This suppresses the incomes of the very wealthy and causes the Census Gini to understate true income inequality, particularly at the upper tail. The World Inequality Database and Piketty-Saez series, which draw on tax records, do not share this limitation.
Known Limitations
  • Excludes capital gains, which are highly concentrated at the top; substantially understates top-end inequality
  • Excludes non-cash transfers (SNAP, Medicaid, housing vouchers), understating the redistributive effect of government programs
  • Based on self-reported household survey data; subject to non-response bias and misreporting
  • Unit of analysis is the household, not the individual or tax unit; household size has fallen over time
  • Three documented series breaks (1993, 2013-14, 2017-18) complicate long-run trend analysis

Series 3

Official Poverty Rate (OPM)

Coverage: 1959–2024  |  Updated: September 2025  |  Unit: Persons  |  Frequency: Annual
Primary Citation
Emily A. Shrider, “Poverty in the United States: 2024,” Current Population Reports P60-287 (U.S. Census Bureau, September 2025). Earlier years from Bernadette D. Proctor, Jessica L. Semega, and Melissa A. Kollar, and prior annual poverty reports in the P60 series.
What Is Measured
The share of persons in the U.S. civilian non-institutionalized population with pre-tax cash income below the applicable Orshansky poverty threshold. Thresholds vary by family size and composition; for a family of four in 2024, the threshold is approximately $32,150. Income is measured the same way as the Census Gini: pre-tax cash income, excluding capital gains and non-cash transfers.
The Orshansky Thresholds
Mollie Orshansky developed the poverty thresholds in 1963–64 at the Social Security Administration, based on the cost of the USDA Economy Food Plan multiplied by three — reflecting survey data showing that families spent approximately one-third of their budgets on food in 1955. The thresholds have been updated annually for CPI inflation since 1969 but have never been revised to reflect changes in consumption patterns, living standards, or geographic cost-of-living variation. The income share of food is now approximately one-eighth rather than one-third of family budgets, suggesting the thresholds are severely outdated as a measure of material deprivation.
What Is Not Counted as Income
The OPM counts only pre-tax cash income. The following are explicitly excluded:
  • SNAP (food stamps) — the largest in-kind food assistance program
  • Medicaid and Medicare benefits
  • Housing vouchers (Section 8) and public housing subsidies
  • The Earned Income Tax Credit (EITC), which provides up to ~$7,800 to low-income working families
  • The Child Tax Credit refundable portion
  • School meals programs
Because these programs have expanded dramatically since the 1960s, the OPM systematically understates the reduction in material deprivation achieved by government programs. Under the OPM, expanding SNAP has zero effect on measured poverty.
Supplemental Poverty Measure
The Supplemental Poverty Measure (SPM), published annually by the Census Bureau since 2011 (with experimental series back to 2009), is the preferred modern alternative. The SPM uses updated thresholds based on expenditures on food, clothing, shelter, and utilities (FCSU); counts all government transfers including SNAP and the EITC; deducts taxes paid, work expenses, and out-of-pocket medical costs; and adjusts for geographic cost-of-living variation. SPM poverty rates differ substantially from OPM rates and tell a different story about which programs reduce poverty. Readers conducting policy analysis should consult both measures.
Known Limitations
  • Thresholds not updated for changing living standards since 1969; widely regarded as anachronistic
  • Excludes nearly all major anti-poverty programs from income count
  • No geographic variation in thresholds; same threshold in rural Mississippi and San Francisco
  • Based on self-reported CPS survey income; subject to underreporting of transfer income
  • Institutionalized population (prisons, nursing homes) excluded; incarceration rate has risen dramatically since the 1970s

Series 4 — UNOFFICIAL

Pre-1959 Poverty Estimates

Coverage: 1914–1958  |  Status: Unofficial — not suitable for quantitative analysis
Critical Warning: No household survey capable of supporting official poverty estimates existed before 1947. All pre-1959 poverty figures are econometric backcasts with very high uncertainty. One set of estimates was withdrawn by its author. These figures are displayed for illustrative historical context only and must be labeled as unofficial on all charts.
Source 1 (1947–1958)
Gordon M. Fisher, “Estimates of the Poverty Population Under the Current Official Definition for Years Before 1959” (ASPE/U.S. Department of Health and Human Services, 1986). These estimates were withdrawn by Fisher himself in 1999. Fisher concluded that the methodological assumptions underlying the backcast were not defensible, and explicitly asked that the figures not be cited or used.
Source 2 (1914–1946)
Robert D. Plotnick, Eugene Smolensky, Eirik Evenhouse, and Siobhan Reilly, “The Twentieth-Century Record of Inequality and Poverty in the United States,” in Stanley L. Engerman and Robert E. Gallman, eds., The Cambridge Economic History of the United States, vol. III: The Twentieth Century (Cambridge University Press, 2000), pp. 249–300. The authors themselves caution: “It strikes us as unreasonable to assert that 60% of Americans were poor in 1920” — illustrating the extreme sensitivity of the figures to methodological assumptions about how to backdate Orshansky thresholds to periods when consumption baskets, family structures, and living standards were dramatically different.
Why No Survey Existed
The Current Population Survey was established by the Census Bureau and Bureau of Labor Statistics in 1940 and has been conducted monthly since 1947. Before 1947, no nationally representative household income survey existed. Income data from this period comes from decennial censuses (which did not measure income until 1940), tax records (which covered only a small fraction of the population in the early 20th century), and various employer surveys. The Orshansky methodology cannot be applied to pre-survey data without compounding assumptions that Plotnick et al. acknowledge are potentially unreasonable.

Citation

How to Cite This Platform

Suggested Citation

Equally Poor: Visualizing Wealth, Inequality & Law in America (2025). Interactive data platform. Available at: [URL]. Data sourced from Piketty & Saez (2024 update); U.S. Census Bureau CPS ASEC (FRED: GINIALLRH); Shrider, Poverty in the United States: 2024, P60-287 (2025).

When citing specific data series, please cite the primary sources directly — see citations above for each series.

Annotated Bibliography

Further Reading

Key works in the empirical and theoretical literature on inequality, poverty, and law.

  • Piketty, Thomas. Capital in the Twenty-First Century. Translated by Arthur Goldhammer. Cambridge, MA: Belknap Press of Harvard University Press, 2014.

    The landmark synthetic work arguing that the rate of return on capital (r) systematically exceeds economic growth (g) over the long run, driving wealth concentration absent countervailing forces such as progressive taxation or war. Draws on Piketty-Saez data and longer-run wealth series for France, Britain, and Germany. Essential starting point; see also critical responses by Acemoglu & Robinson, Krusell & Smith, and others.

  • Piketty, Thomas, and Emmanuel Saez. “Income Inequality in the United States, 1913–1998.” Quarterly Journal of Economics 118, no. 1 (2003): 1–41.

    The foundational paper constructing the U.S. top income share series from IRS Statistics of Income data. Methodological appendix is essential reading for understanding what the series measures and does not measure.

  • Auten, Gerald, and David Splinter. “Income Inequality in the United States: Using Tax Data to Measure Long-Term Trends.” Journal of Political Economy 132, no. 7 (2024): 2179–2227.

    The most comprehensive challenge to the Piketty-Saez findings, arguing that after adjusting for taxes, transfers, and non-filers, post-tax-and-transfer inequality has increased far less than the pre-tax series suggests. Required reading alongside Piketty-Saez for any policy-oriented analysis.

  • Chetty, Raj, Nathaniel Hendren, Patrick Kline, and Emmanuel Saez. “Where is the Land of Opportunity? The Geography of Intergenerational Mobility in the United States.” Quarterly Journal of Economics 129, no. 4 (2014): 1553–1623.

    Uses administrative tax records linked across generations to document dramatic geographic variation in intergenerational mobility. Finds that some U.S. metros have mobility rates comparable to Denmark while others are far below the U.S. average. Connects inequality to opportunity in legally and politically tractable ways.

  • Atkinson, Anthony B. Inequality: What Can Be Done? Cambridge, MA: Harvard University Press, 2015.

    A comprehensive normative and empirical treatment of inequality by one of the founders of the modern measurement literature. Proposes fifteen specific policy interventions; particularly valuable for legal scholars because Atkinson takes institutional design seriously.

  • Card, David, and Alan B. Krueger. “Minimum Wages and Employment: A Case Study of the Fast-Food Industry in New Jersey and Pennsylvania.” American Economic Review 84, no. 4 (1994): 772–793.

    The landmark natural experiment showing that New Jersey's 1992 minimum wage increase did not reduce fast-food employment relative to neighboring Pennsylvania. Opened the modern empirical literature on minimum wages and helped rehabilitate labor market regulation as a distributional tool.

  • Dube, Arindrajit. “Minimum Wages and the Distribution of Family Incomes.” American Economic Journal: Applied Economics 11, no. 4 (2019): 268–304.

    Uses cross-state variation in minimum wages to show that higher minimum wages substantially reduce poverty rates and compress the lower tail of the income distribution. Directly relevant to the legal question of what labor law can accomplish as a distributional instrument.

  • Hacker, Jacob S., and Paul Pierson. Winner-Take-All Politics: How Washington Made the Rich Richer — and Turned Its Back on the Middle Class. New York: Simon & Schuster, 2010.

    Political science account of how organized business interests systematically shaped tax law, financial regulation, and labor law over four decades to produce the observed rise in top income shares. Essential bridge between the economics literature and legal-institutional analysis.

  • Saez, Emmanuel, and Gabriel Zucman. The Triumph of Injustice: How the Rich Dodge Taxes and How to Make Them Pay. New York: W.W. Norton, 2019.

    Synthesizes the authors' research on effective tax rates across the full income and wealth distribution, arguing that the U.S. tax system has become regressive at the very top. Includes simulation of alternative tax structures. Essential for tax law scholars.

  • Goldin, Claudia, and Lawrence F. Katz. The Race Between Education and Technology. Cambridge, MA: Belknap Press of Harvard University Press, 2008.

    Attributes the rise in wage inequality primarily to skill-biased technological change outpacing educational attainment. An important complement to the legal-political accounts; any adequate explanation of rising inequality must engage with both technological and institutional factors.

  • Mishel, Lawrence, Josh Bivens, Elise Gould, and Heidi Shierholz. The State of Working America, 12th ed. Ithaca, NY: Cornell University Press / Economic Policy Institute, 2012.

    Comprehensive reference volume on wages, income, wealth, poverty, and mobility in the United States. Particularly valuable for its long-run wage series and documentation of divergence between productivity and compensation since the 1970s.

  • Alstott, Anne L., and Bruce Ackerman. The Stakeholder Society. New Haven: Yale University Press, 1999.

    Legal-normative argument for a universal capital grant as a mechanism for equalizing opportunity. A leading example of how legal scholars have engaged with the empirical inequality literature to develop institutional proposals.

  • Shrider, Emily A. “Poverty in the United States: 2024.” Current Population Reports P60-287. Washington, DC: U.S. Census Bureau, September 2025.

    The official annual poverty report for reference year 2024. Contains both Official Poverty Measure and Supplemental Poverty Measure estimates with full demographic breakdowns.