Part 1 of this series defined a market society. A market society is the arrangement in which price alone settles the allocation of land and labor (Part 1). Part 2 followed that arrangement to the colonies (Part 2). Part 3 examined the same arrangement and the atmosphere (Part 3). This post examines the same arrangement and the household budget.

Four terms do the work here. I use concentration for the rise in the share of income and wealth that the top of the distribution holds. I use precarity for work that gives no security of hours, income, or tenure. I use affordability crisis for the condition in which housing, healthcare, and education cost more than ordinary households can pay. I use labor market institutions for the rules that set pay and bargaining power. Those rules include union contracts, minimum wage law, and the law of employment.

That last definition is deliberately narrow. A wider definition covers every rule that touches a market, including land use law and student loan law. A wider definition also makes the argument of this post impossible to test. The addition of a rule and the removal of a rule then confirm it equally well. A section below presents three prices that this narrow frame does not explain.

What this post claims

This post makes three claims of different strength.

1. Income and wealth concentrated at the top of the United States distribution after about 1979. Scholars dispute the size of the rise and agree on its direction.
2. Changes in labor market institutions explain a large minority of the rise in wage inequality. The best estimate is a fifth to a third.
3. Ethnography shows how households live the result. This claim describes mechanisms and meanings rather than magnitudes.

Claim 2 is the main claim, and I state its limit now. No comparable decomposition exists for the top 0.1% wealth share, which moves with business income, capital gains, and asset prices. Union decline is close to silent about that share. When this post says that institutions explain concentration, it means wage dispersion among workers. That is one part of concentration, and the other parts need other explanations.

The strongest objections

Two bodies of work dispute claim 2. The first disputes the size of the target. The second disputes the cause.

Gerald Auten and David Splinter, both economists, rebuilt the tax-based estimates and reached smaller numbers (Auten & Splinter, 2024a). Their result is smaller before tax and after tax. They estimate that the top 0.1% pre-tax income share moved from 3.2% to 5.4% between 1979 and 2019. Thomas Piketty, Emmanuel Saez, and Gabriel Zucman, all economists, report a much larger movement. Auten and Splinter also estimate that the top 1% share of income after taxes and transfers rose about 1.4 percentage points since 1979.

I cannot settle that dispute here, and it does not turn on taxes alone. The disagreement is about how each team assigns income that nobody reports on a tax return, such as untaxed business income and untaxed capital income. Piketty, Saez, and Zucman dispute that assignment (Piketty, Saez, & Zucman, 2024). They argue that the evidence on tax evasion does not support it. Auten and Splinter answered that comment in turn (Auten & Splinter, 2024b). Both teams find a rise, and they disagree about its size by a factor of two or more.

The second dispute is about cause. Claudia Goldin and Lawrence Katz, both economists, explain the wage structure as a race (Goldin & Katz, 2008). The race runs between the demand for skill and the supply of skill. Technology raises the demand for educated workers, and education supplies them at a rate that slowed after 1980. In their account the wage gap widens because the supply grew too slowly. No change in union membership is necessary to produce it.

Loukas Karabarbounis and Brent Neiman, both economists, dispute the financial mechanism directly (Karabarbounis & Neiman, 2014). Labor’s share of income fell in most countries and in most industries. The relative price of investment goods fell, and that fall pushed firms to buy machines instead of hours. They attribute about half of the drop in labor’s share to it. A cause that operates in countries with intact union systems cannot be a cause specific to deregulation in the United States.

Two things survive these objections.

• The skill explanation and the institutional explanation are compatible. Bruce Western and Jake Rosenfeld, both sociologists, estimate that union decline explains a fifth to a third of the growth in wage inequality. They describe that effect as comparable in size to the effect of education (Western & Rosenfeld, 2011).
• Karabarbounis and Neiman constrain the size of the institutional claim without removing it. A global cause and a national cause can operate together on the same quantity.

The defensible version of claim 2 is therefore a minority share of one part of concentration.

The scale

Every number below is a pre-tax number unless the text says otherwise.

• Saez and Zucman report that the top 1% held 38% of household wealth in 2018 (Saez & Zucman, 2020).
• The same paper reports that the bottom 50% earned 12.5% of pretax national income in 2018.
• The same paper reports that the top 0.1% wealth share rose from about 7% in 1978 to about 18% in 2018.
• Carter Price, a mathematician, and Kathryn Edwards, a labor economist, hold the 1975 distribution of real taxable income fixed (Price & Edwards, 2020). They grow every income at the growth rate of real per capita gross domestic product, which was 118%. The gap between that calculation and actual incomes below the 90th percentile totals $47 trillion from 1975 to 2018.
• The Economic Policy Institute measures chief executive pay at the 350 largest public firms (Economic Policy Institute, 2025a). That pay reached 281 times the pay of a typical worker in 2024. The same ratio was 31 times in 1978.
• The same institute reports that net productivity grew 90.2% from 1979 to 2025 while pay for a typical worker grew 33.0% (Economic Policy Institute, 2025b).

Each of those numbers carries a qualification, and the qualifications belong next to the numbers.

The wealth shares come from a contested method. Saez and Zucman revised their own earlier top 0.1% figure downward. Matthew Smith, Owen Zidar, and Eric Zwick, all economists, use a method that allows different investors to earn different returns. They put the top 0.1% share at 15.7% in 2016 (Smith, Zidar, & Zwick, 2023). The range across careful teams runs from about 15% to about 20%.

The $47 trillion total comes from a RAND working paper that RAND did not send through peer review. The calculation grows all incomes at the rate of per capita output. It therefore produces a gap whenever income growth trails output growth for any reason. Changes in the composition of households are one such reason.

The $47 trillion total is also a pre-tax and pre-transfer figure. Taxes and transfers reduce it, and this post cannot report the reduced figure because the source does not calculate one. Some press coverage rounded the total up to $50 trillion, which the source does not support.

The chief executive figure uses realized pay, which counts stock options at the moment the executive exercises them. That measure follows the stock market closely. The Economic Policy Institute publishes a granted-pay series that rises far less. Steven Kaplan and Joshua Rauh, both economists, offer a different account (Kaplan & Rauh, 2013). They argue that pay at the top rose with firm size and with the market for a small number of executives.

The productivity gap has a direct rebuttal. Anna Stansbury and Lawrence Summers, both economists, regress pay growth on productivity growth from 1973 to 2016 (Stansbury & Summers, 2018). They find that one point of higher productivity growth still comes with 0.7 to 1.0 points of higher median and average compensation growth. The figure for production and nonsupervisory workers is 0.4 to 0.7 points. On their reading the link between pay and productivity holds, and other forces push typical pay down at the same time.

One more caution applies to the whole section. The $47 trillion figure treats the years from 1945 to 1974 as the standard. A later section accepts an anthropological objection that secure and equal employment was itself brief and local. I cannot hold both positions without a qualification, so here is the qualification. The postwar distribution came from war controls, closed immigration, and a unique position in the world economy. A calculation built on it measures the distance from an unusual period.

How the rules changed

Union membership fell across the whole period. Western and Rosenfeld report the fall in private sector union membership between 1973 and 2007 (Western & Rosenfeld, 2011). It fell from 34% to 8% for men and from 16% to 6% for women. The peak came in the middle 1950s, so the fall overlaps the growth of financial income instead of preceding it.

Their argument is about norms as well as about contracts. They find lower pay dispersion in regions and industries with high union membership, including among workers with no union contract. They read that pattern as the effect of a norm. The reading is an inference from an association. It is the most contested part of their paper, and I use it as a proposed mechanism.

Finance grew over the same decades. Ken-Hou Lin and Donald Tomaskovic-Devey, both sociologists, measure how much of a nonfinancial firm’s income comes through financial channels rather than through production. Their counterfactual analysis covers 1970 to 2008 (Lin & Tomaskovic-Devey, 2013). It attributes more than half of the decline in labor’s share of income to that shift. It also attributes 9.6% of the growth in the officers’ share of compensation, and 10.2% of the growth in earnings dispersion.

Their claim is that a firm which earns through financial channels depends less on its own workers. The workers then lose bargaining power. Karabarbounis and Neiman give the reason to hold that claim loosely. A decomposition attributes a share of a trend under a set of assumptions. It does not exclude a global cause that operates on the same trend.

Karen Ho, an anthropologist, supplies the mechanism that the statistics cannot supply. She worked at a Wall Street investment bank and interviewed more than one hundred people in the industry (Ho, 2009).

Her bankers valued liquidity, and they applied that value to their own careers and to the firms they advised. They lost jobs often, and they treated the loss as normal. They carried the same expectation into their advice. Job loss looked like discipline to them rather than damage. Ho shows how shareholder value moved from a doctrine into ordinary practice. Part 3 traced the same doctrine into the oil industry.

How households live it

Katherine Newman, an anthropologist, interviewed managers, engineers, and other middle class people who lost their position and could not regain it (Newman, 1988/1999). Her subjects explained the fall as a personal failure. The culture gave them a story in which effort produces reward, so the loss of reward implied a lack of effort. They hid the fall from neighbors and sometimes from family.

Kathryn Marie Dudley, an anthropologist, studied Kenosha, Wisconsin after Chrysler closed the assembly plant in 1988 (Dudley, 1994). She found a town that argued with itself about what the closure meant. Professional residents read the closure as the arrival of a better economy that rewards education. The autoworkers read it as the removal of a way of life that they earned. The disagreement was about moral standing rather than about the plant.

Anna Tsing, an anthropologist, describes how a supply chain pushes risk outward. A lead firm sets terms and takes the margin, and the firms below it absorb the variation in demand (Tsing, 2009). She argues that the chain recruits labor through difference. Gender, nationality, religion, and immigration status become tools of that recruitment. The chain sorts workers by those differences and prices the differences into the contract.

Anthropology also argues with itself about the word precarity. Clara Han, an anthropologist, reviews the literature and warns that the word compresses two different things (Han, 2018). One is a historical condition of insecure employment. The other is a general human vulnerability that no economic system removes.

Kathleen Millar, an anthropologist who publishes in a sociology journal, makes a related objection (Millar, 2017). She notes that secure wage employment was itself brief and local. That assumption makes most of the world’s workers an exception. I accept the objection, and the section above applies it to my own largest figure. The claim I defend is that secure employment became less common in the United States after 1979.

Debt

David Graeber, an anthropologist, argues that debt begins as a moral relation and becomes a quantity (Graeber, 2011). An obligation between people is open and negotiable. Money makes it exact, transferable, and enforceable. His historical claims drew serious criticism from specialists, and Part 1 noted that criticism. The narrow point survives it, because the moral force of a debt is visible in the fieldwork whatever the origin story.

Gustav Peebles, an anthropologist, reviews the field and states the point that this post needs (Peebles, 2010). Credit and debt are one relation seen from two sides. A lender calls the contract an asset, and a borrower calls the same contract a burden. The moral language attaches to the borrower.

Deborah James, an anthropologist, studied borrowing in South Africa after apartheid (James, 2015). The state extended credit to black South Africans as a form of economic inclusion. Her subjects used the credit to fund status and mobility, such as houses, funerals, and school fees. Many of them ended with debt that they could not service, and a private advice industry grew to manage them. One national case shows that the pattern reaches past the United States. One national case does not show where else it appears.

Three sectors, and three prices this frame does not explain

Here the argument of this post meets its hardest test. In each of the three sectors, a serious literature traces the price to a cause outside my frame. Two of those causes are added rules. The third is concentrated market power.

Housing is the largest item. The Harvard Joint Center for Housing Studies counts cost-burdened renter households (Joint Center for Housing Studies, 2025). It reports 22.6 million of them in 2023, which is half of all renters. Matthew Desmond, a sociologist who uses ethnographic method, followed eight Milwaukee families through eviction (Desmond, 2016). Desmond and Nathan Wilmers later measured the pattern that the ethnography suggested. They report that landlords in poor neighborhoods overcharge relative to property value and take higher profits (Desmond & Wilmers, 2019).

Edward Glaeser and Joseph Gyourko, both economists, examine the gap between house prices and construction costs (Glaeser & Gyourko, 2018). They argue that the gap comes from land use regulation rather than from a physical shortage of land. If they are right, then added regulation is a cause of the housing crisis.

Education shows the same shape. David Lucca, Taylor Nadauld, and Karen Shen, all economists, measure what happens when federal loan caps rise (Lucca, Nadauld, & Shen, 2019). They find that institutions pass about 60 cents of each subsidized loan dollar into tuition, and about 20 cents of each unsubsidized dollar. Expanded public credit raised the price that the credit was meant to help students pay.

Healthcare shows a third version. Zack Cooper, Stuart Craig, Martin Gaynor, and John Van Reenen are all economists. They analyze claims data for 28% of people with employer-sponsored insurance (Cooper et al., 2019). They find that hospital prices in monopoly markets run 15.3% above prices in competitive markets. That is a finding about market structure and about weak enforcement of competition.

My frame does not explain these three prices, and I have no version of it that does. The most I can defend is that two mechanisms operate at once. Restricted supply and concentrated market power raise a price. Stagnant pay at the bottom of the distribution decides who absorbs the increase. Glaeser and Gyourko, Lucca and colleagues, and Cooper and colleagues document the first half in all three sectors. Desmond and Wilmers document the second half in housing alone.

The rest of the sector evidence is a set of totals rather than a set of causes.

Healthcare debt is wide. KFF Health News surveyed a national sample in 2022 (KFF Health News, 2022). It estimated that more than 100 million adults in the United States carry healthcare debt, which is 41% of adults.

David Himmelstein and colleagues surveyed 910 people who filed for personal bankruptcy between 2013 and 2016. They report that 58.5% of debtors cited medical bills as a contributor (Himmelstein et al., 2019). They report that 66.5% cited some medical contributor. Scholars dispute the causal share. David Dranove and Michael Millenson reanalyzed the 2001 data behind an earlier Himmelstein study and put the contribution near 17% (Dranove & Millenson, 2006). A cited contributor is not a proven cause, and the two figures measure different things.

Education debt is deep, and the price signal points the other way. Federal student loans total about $1.67 trillion across about 42 million borrowers (Federal Student Aid, 2025). Published tuition at public four-year institutions doubled in inflation-adjusted terms between 1995 and 2025. Average inflation-adjusted net tuition and fees at those institutions peaked at $4,450 in 2012 and fell to about $2,300 in 2025 (College Board, 2025).

The net figure is the strongest fact against my own education case, and it comes from a source that this post already cites. The debt stock reflects three decades of accumulated borrowing and a shift of cost onto students. The current net price at a public four-year institution falls. I therefore withdraw education from the affordability claim of this post and keep it as a debt claim.

The counterexamples this frame has to face

Two bodies of work challenge the whole shape of the argument.

Timothy Kohler and colleagues measure wealth inequality from the distribution of house sizes at 63 archaeological sites (Kohler et al., 2017). Inequality rises with domestication and with the scale of political organization, thousands of years before any market society. Concentration is therefore not a signature of the arrangement this series studies. The claim I can defend is about a rise inside one country in one period.

Walter Scheidel, a historian, examines the periods when inequality fell (Scheidel, 2017). He finds four causes that recur, which are mass mobilization war, revolution, state collapse, and plague. On his account the postwar compression came from the world wars instead of from a choice of policy. If he is right, then the rules of the market matter less than this post assumes, and they are also harder to change back. I have no refutation of Scheidel. His own account still gives policy a role in the decades after each shock.

Claims that need evidence

I list the causal claims that this post asserts without proving them.

• Union norms set pay in nonunion workplaces. Western and Rosenfeld infer this from a regional and industry association.
• Financialization reduced worker bargaining power. Lin and Tomaskovic-Devey show a decomposition under assumptions, and Karabarbounis and Neiman offer a rival account of the same trend.
• Wall Street culture spread downsizing through the wider economy. Ho documents the culture inside the banks. She does not measure how much downsizing it produced elsewhere.
• Restricted supply raises the return to owning existing housing. This is a standard price argument, and this post cites no measurement of it.
• Debt disciplines borrowers into compliant behavior. The ethnography shows that borrowers feel the moral weight. It does not measure how their political or workplace behavior changes.
• Concentration at the top caused the affordability crisis in housing and healthcare. The two trends run together in time, and the section above gives a rival cause for each price.
• Stagnant pay decides who absorbs a price increase. Desmond and Wilmers show this for rental housing. No source here shows it for tuition or for hospital prices.
• A market society converts an institutional outcome into a personal verdict. Three ethnographies in two countries show the pattern. They do not establish it for market societies in general, and the last section gives the observation that would refute it.

What survives, and what would falsify it

The rules that set pay and bargaining power in the United States changed after about 1979. Union decline explains a fifth to a third of the rise in wage dispersion. That is a minority share of one part of concentration. Scholars actively dispute the measured size of concentration itself. The affordability crisis in housing and healthcare has causes that this frame does not supply.

Ethnography adds what the aggregates omit, which is the way people explain the result to themselves. Newman’s subjects blamed their own effort. Dudley’s town argued about who deserved the loss. James’s borrowers read credit as a route to standing. The pattern is that a market society converts an institutional outcome into a personal verdict.

That last claim is a real claim, so it needs a falsifier. It fails wherever people who suffer the same loss attribute it outward, and then organize on that basis. They can attribute the loss to government, to employers, or to a collective wrong. Such cases exist in the same literature, and Part 5 of this series takes them as its subject. Outward attribution may turn out to be the common case and self-blame the exception. In that event the claim in this post is wrong and the claim in the next post is right.

The final post asks what pushes back.

References

Auten, G., & Splinter, D. (2024a). Income inequality in the United States: Using tax data to measure long-term trends. Journal of Political Economy, 132(7), 2179-2227. https://doi.org/10.1086/728741

Auten, G., & Splinter, D. (2024b). Reply to Piketty, Saez, and Zucman (2024): Income inequality in the United States: A comment. https://www.davidsplinter.com/AutenSplinter-2024-ReplyToPSZ.pdf

College Board. (2025). Trends in college pricing and student aid 2025. https://research.collegeboard.org/trends/college-pricing/highlights

Cooper, Z., Craig, S. V., Gaynor, M., & Van Reenen, J. (2019). The price ain’t right? Hospital prices and health spending on the privately insured. The Quarterly Journal of Economics, 134(1), 51-107. https://doi.org/10.1093/qje/qjy020

Desmond, M. (2016). Evicted: Poverty and profit in the American city. Crown. https://www.penguinrandomhouse.com/books/247816/evicted-by-matthew-desmond/

Desmond, M., & Wilmers, N. (2019). Do the poor pay more for housing? Exploitation, profit, and risk in rental markets. American Journal of Sociology, 124(4), 1090-1124. https://doi.org/10.1086/701697

Dranove, D., & Millenson, M. L. (2006). Medical bankruptcy: Myth versus fact. Health Affairs, 25(2), W74-W83. https://doi.org/10.1377/hlthaff.25.w74

Dudley, K. M. (1994). The end of the line: Lost jobs, new lives in postindustrial America. University of Chicago Press. https://press.uchicago.edu/ucp/books/book/chicago/E/bo3618730.html

Economic Policy Institute. (2025a). CEO pay. https://www.epi.org/publication/ceo-pay/

Economic Policy Institute. (2025b). The productivity-pay gap. https://www.epi.org/productivity-pay-gap/

Federal Student Aid. (2025). Federal student loan portfolio. U.S. Department of Education. https://studentaid.gov/data-center/student/portfolio

Glaeser, E., & Gyourko, J. (2018). The economic implications of housing supply. Journal of Economic Perspectives, 32(1), 3-30. https://doi.org/10.1257/jep.32.1.3

Goldin, C., & Katz, L. F. (2008). The race between education and technology. Harvard University Press. https://www.hup.harvard.edu/books/9780674035300

Graeber, D. (2011). Debt: The first 5,000 years. Melville House. https://davidgraeber.org/books/debt-the-first-5000-years/

Han, C. (2018). Precarity, precariousness, and vulnerability. Annual Review of Anthropology, 47, 331-343. https://doi.org/10.1146/annurev-anthro-102116-041644

Himmelstein, D. U., Lawless, R. M., Thorne, D., Foohey, P., & Woolhandler, S. (2019). Medical bankruptcy: Still common despite the Affordable Care Act. American Journal of Public Health, 109(3), 431-433. https://doi.org/10.2105/AJPH.2018.304901

Ho, K. (2009). Liquidated: An ethnography of Wall Street. Duke University Press. https://www.dukeupress.edu/liquidated

James, D. (2015). Money from nothing: Indebtedness and aspiration in South Africa. Stanford University Press. https://www.sup.org/books/anthropology/money-nothing

Joint Center for Housing Studies of Harvard University. (2025). The state of the nation’s housing 2025. https://www.jchs.harvard.edu/state-nations-housing-2025

Kaplan, S. N., & Rauh, J. (2013). It’s the market: The broad-based rise in the return to top talent. Journal of Economic Perspectives, 27(3), 35-56. https://doi.org/10.1257/jep.27.3.35

Karabarbounis, L., & Neiman, B. (2014). The global decline of the labor share. The Quarterly Journal of Economics, 129(1), 61-103. https://doi.org/10.1093/qje/qjt032

KFF Health News. (2022, June 16). 100 million people in America are saddled with health care debt. https://kffhealthnews.org/news/article/diagnosis-debt-investigation-100-million-americans-hidden-medical-debt/

Kohler, T. A., Smith, M. E., Bogaard, A., Feinman, G. M., Peterson, C. E., Betzenhauser, A., Pailes, M., Stone, E. C., Prentiss, A. M., Dennehy, T. J., Ellyson, L. J., Nicholas, L. M., Faulseit, R. K., Styring, A., Whitlam, J., Fochesato, M., Foor, T. A., & Bowles, S. (2017). Greater post-Neolithic wealth disparities in Eurasia than in North America and Mesoamerica. Nature, 551, 619-622. https://doi.org/10.1038/nature24646

Lin, K.-H., & Tomaskovic-Devey, D. (2013). Financialization and U.S. income inequality, 1970-2008. American Journal of Sociology, 118(5), 1284-1329. https://doi.org/10.1086/669499

Lucca, D. O., Nadauld, T., & Shen, K. (2019). Credit supply and the rise in college tuition: Evidence from the expansion in federal student aid programs. The Review of Financial Studies, 32(2), 423-466. https://doi.org/10.1093/rfs/hhy069

Millar, K. M. (2017). Toward a critical politics of precarity. Sociology Compass, 11(6), e12483. https://doi.org/10.1111/soc4.12483

Newman, K. S. (1999). Falling from grace: Downward mobility in the age of affluence. University of California Press. (Original work published 1988 by Free Press.) https://www.ucpress.edu/books/falling-from-grace/paper

Peebles, G. (2010). The anthropology of credit and debt. Annual Review of Anthropology, 39, 225-240. https://doi.org/10.1146/annurev-anthro-090109-133856

Piketty, T., Saez, E., & Zucman, G. (2024). Income inequality in the United States: A comment (World Inequality Lab Technical Note No. 2024/04). https://gabriel-zucman.eu/files/PSZ2024.pdf

Price, C. C., & Edwards, K. A. (2020). Trends in income from 1975 to 2018 (Working Paper WR-A516-1). RAND Corporation. https://www.rand.org/pubs/working_papers/WRA516-1.html

Saez, E., & Zucman, G. (2020). The rise of income and wealth inequality in America: Evidence from distributional macroeconomic accounts. Journal of Economic Perspectives, 34(4), 3-26. https://doi.org/10.1257/jep.34.4.3

Scheidel, W. (2017). The great leveler: Violence and the history of inequality from the Stone Age to the twenty-first century. Princeton University Press. https://press.princeton.edu/books/paperback/9780691183251/the-great-leveler

Smith, M., Zidar, O., & Zwick, E. (2023). Top wealth in America: New estimates under heterogeneous returns. The Quarterly Journal of Economics, 138(1), 515-573. https://doi.org/10.1093/qje/qjac033

Stansbury, A., & Summers, L. H. (2018). Productivity and pay: Is the link broken? (NBER Working Paper No. 24165). National Bureau of Economic Research. https://www.nber.org/papers/w24165

Tsing, A. (2009). Supply chains and the human condition. Rethinking Marxism, 21(2), 148-176. https://doi.org/10.1080/08935690902743088

Western, B., & Rosenfeld, J. (2011). Unions, norms, and the rise in U.S. wage inequality. American Sociological Review, 76(4), 513-537. https://doi.org/10.1177/0003122411414817