GDP, household stress, and the illusion of prosperity — how forty years of measured growth produced an economy in which one in three Americans cannot cover a four-hundred-dollar emergency, and what that decoupling was built out of.
For most of the postwar period in the United States, productivity growth and the compensation of typical workers tracked one another. Between 1948 and 1973, output per hour of labor in the U.S. economy rose by approximately ninety-five percent. Hourly compensation of production and nonsupervisory workers — the roughly eighty percent of the workforce that did most of the producing — rose by approximately ninety percent in the same period. The relationship was not coincidental. It was the operating signature of an economic settlement in which workers’ organized power, public investment, and corporate practice each contributed to channeling productivity gains into worker compensation. The settlement was contested, partial, and unequally distributed by race and gender, but its aggregate signature was clear: when the economy grew, the people who did the work grew with it.
Beginning around 1973, the relationship broke. From 1973 to 2023, U.S. productivity rose by approximately seventy-five percent. Compensation for the same eighty percent of the workforce rose by approximately fifteen percent. The Economic Policy Institute, which has tracked the divergence in its State of Working America data since the 1980s, calls this the productivity-pay gap. It is the hidden subject of every U.S. economic conversation that treats GDP growth as a description of national prosperity. Growth has continued. The relationship between growth and household security has not.
This article is about that gap. Begin with the question almost no one in mainstream economic coverage asks: what is prosperity?
The dominant U.S. answer is aggregate output growth. Real GDP growth of two-to-three percent annually is described, in political speech and in financial press headlines, as a strong economy. Per-capita GDP growth is treated as direct evidence of rising living standards. Productivity growth is treated as the mechanism by which aggregate prosperity reaches the population. When the headline indicators are positive, the economy is described as doing well, and the public is understood to be living well within it.
A different answer is operationally relevant. From a household perspective, prosperity is not the rate of aggregate output growth. It is the absence of the recurring fear that one bad event — a medical emergency, a car breakdown, a layoff, a rent increase, a child’s illness, an aging parent’s decline — will collapse the household’s economic standing. It is the presence of a buffer between routine economic life and economic ruin. It is the structural assumption that the household will be able to recover from a shock without having to liquidate, to borrow, to cut essential consumption, or to make the kinds of decisions the polite phrase trade-offs obscures the severity of. Prosperity, in this sense, is economic security at the household level. It is a property of arrangements rather than of indicators.
The two definitions of prosperity have been decoupled in the United States for forty years. The decoupling is not a discovery of economics. It is the operational consequence of a set of institutional changes — in labor markets, in healthcare, in housing, in retirement, in childcare, in education, in time arrangements, in regulatory architecture — that have systematically transferred the burden of economic risk from firms and from public institutions onto household balance sheets. The aggregate growth has continued. The security has not.
The Federal Reserve, in its annual Survey of Household Economics and Decisionmaking, has documented one component of the resulting precarity for over a decade. In the 2023 survey, approximately thirty-seven percent of U.S. adults reported they would be unable to cover an unexpected four-hundred-dollar expense entirely with cash or its equivalent. The figure has been roughly stable since the survey began in 2013. The percentage of U.S. adults experiencing this level of liquidity precarity in the world’s wealthiest economy, in a period when GDP per capita has reached unprecedented levels, in years officially described in political speech as a strong economy, has been approximately one in three for a decade.
Household absorption is the process by which firms with pricing or labor-market power transfer the costs and risks of economic activity onto household balance sheets, while retaining the productivity gains and the revenue. The mechanism does not require malice. It requires only an institutional arrangement in which the costs of healthcare, housing, education, childcare, retirement security, and economic shock are borne primarily by individual households rather than pooled across the broader economy through public institutions, employer obligations, or organized labor’s bargained protections.
It rarely operates alone. Its dominant pairing is household absorption plus wage suppression, plus cost shifting, plus risk dumping, plus time theft — the family signature of Family 4. This is reinforced upstream by margin defense (Family 1), the mechanism by which firms with pricing power capture productivity gains rather than passing them through, and by asset conversion (Family 3), the mechanism by which housing has been converted from a shelter system into an asset-income system whose costs are absorbed by tenants and prospective buyers. It is reinforced downstream by narrative laundering (Family 7), the mechanism by which aggregate growth indicators are used to describe an economy in which household security has degraded.
The mechanism’s operational evidence is in the data the federal statistical agencies publish and the financial press largely ignores. The Federal Reserve’s Distributional Financial Accounts document that in 2024, the top ten percent of U.S. households by wealth held approximately seventy percent of all household wealth, while the bottom fifty percent held approximately three percent. The Bureau of Labor Statistics’ Employment Cost Index documents that real compensation growth has been concentrated in the top deciles of the wage distribution for most of the past four decades. The Center for Retirement Research at Boston College documents that approximately half of U.S. households are at risk of being unable to maintain their standard of living in retirement. KFF’s 2022 analysis of survey and credit-bureau data documents that approximately one hundred million Americans carry medical debt. The Joint Center for Housing Studies of Harvard documents that approximately twenty-two million U.S. renter households spend more than thirty percent of their income on rent — the conventional definition of being cost-burdened — and roughly twelve million spend more than half. These are not anecdotes. They are the federal and academic statistical record of household economic life in a period of consistent aggregate growth.
The institutional changes that produced the forty-year decoupling have a documented genealogy. They were not accidental, and they were not produced by anonymous economic forces. They were enacted, court by court, statute by statute, executive order by executive order, contract by contract, by named actors making particular choices.
The first change was the contraction of organized labor’s bargaining power. U.S. union density peaked at roughly thirty-five percent of the private workforce in the mid-1950s. By 1973 it had declined to roughly twenty-five percent. By 2023 it was approximately six percent of the private workforce and approximately ten percent overall. The decline was not driven primarily by workers losing interest in unions; National Labor Relations Board election data and Gallup polling have shown that workers’ interest in union representation has remained substantial throughout the period. The decline was driven by a combination of the National Labor Relations Act’s structural weaknesses (1935), employer-side legal innovation in the 1970s and 1980s, the Reagan administration’s 1981 firing of the Professional Air Traffic Controllers Organization strikers (a moment widely understood by labor historians as a signal to the private sector that the federal government would no longer enforce the postwar bargain), the rightward shift of the federal courts and the National Labor Relations Board, the rise of state-level right-to-work statutes, and the corporate practice of fissuring the workplace through subcontracting, franchising, and gig classification. The result was a workforce in which the institution that had historically translated productivity into compensation had been systematically weakened.
The second change was the shift from defined-benefit retirement to defined-contribution retirement. The Employee Retirement Income Security Act of 1974, the Revenue Act of 1978’s section 401(k), and forty years of subsequent regulatory and tax architecture have transferred the responsibility for retirement provision from employers and pension funds to individual workers managing their own investment accounts. In 1980, approximately sixty percent of U.S. private-sector workers with a workplace retirement plan had a defined-benefit pension; by 2020 the figure was approximately fifteen percent. The shift transferred investment risk, longevity risk, and contribution-decision risk from institutional risk pools to individual households. The Center for Retirement Research’s National Retirement Risk Index estimates that approximately half of U.S. households are now at risk of being unable to maintain their pre-retirement standard of living in retirement. The mechanism is risk dumping (Family 4), and the dumped risk is now permanently distributed across the household sector.
The third change was the transformation of healthcare cost-sharing. Through the 1970s and into the 1980s, the dominant U.S. employer-sponsored health insurance plan structure required low or no out-of-pocket payment for routine care. The shift to managed care in the 1990s, followed by the Health Savings Account and High-Deductible Health Plan architecture introduced in the 2003 Medicare Modernization Act, and accelerated by Affordable Care Act marketplace plan design after 2010, has produced a U.S. health insurance landscape in which the average family deductible exceeds three thousand dollars and the average out-of-pocket maximum exceeds eight thousand dollars. The mechanism is cost shifting (Family 4) — the transfer of routine and emergency healthcare costs from insurers and employers to patients — and its operational signature is the approximately one hundred million Americans who, according to KFF’s 2022 analysis, carry medical debt.
The fourth change was the financialization of housing. Three institutional shifts compounded across forty years. The 1980s deregulation of the savings and loan industry, the 1990s development of mortgage-backed securities markets, and the post-2008 institutional purchase of foreclosed single-family homes by private equity firms transformed U.S. residential housing from a shelter system into an asset class. By 2024, institutional investors owned approximately one in five single-family rentals in some metropolitan areas; the largest single-family rental firms held tens of thousands of homes each. The mechanism is asset conversion (Family 3), and its household-level consequence is the rent burden documented by Harvard’s JCHS: roughly half of all U.S. renter households cost-burdened, with the share at the bottom of the income distribution substantially higher.
The fifth change was the erosion of predictable working time. Just-in-time scheduling, fluctuating hours, on-call shifts, and the algorithmic management of low-wage service work have produced a workforce — particularly in retail, food service, hospitality, and warehouse logistics — in which workers cannot predict their weekly income, cannot arrange childcare reliably, cannot hold a second job to supplement income, and cannot plan around a schedule that is communicated to them with a few days’ notice or less. The mechanism is time theft (Family 4) — not the colloquial sense of workers stealing employer time, but the structural transfer of scheduling risk from firm to household — and its operational signature is the share of the U.S. workforce, estimated by various surveys at roughly twenty to thirty percent, who report being unable to plan more than a week in advance.
The sixth change was the freezing of the federal minimum wage. The federal minimum wage was last raised in July 2009, to $7.25 per hour. As of 2026, it has not been raised in more than sixteen years — the longest stretch without an increase in the statute’s history. Inflation over the same period has eroded its real value by approximately a third. State and municipal minimum wages have substantially exceeded the federal floor in many jurisdictions, but the federal floor remains in effect in roughly twenty states. The mechanism is wage suppression (Family 4), executed through legislative inaction, and its operational signature is a federal minimum wage that, as a fraction of the median wage, is the lowest among OECD economies.
These six changes did not happen in isolation. They reinforced one another. The decline of organized labor weakened the workforce’s ability to resist the cost-shifting, the risk-dumping, the time-theft, and the wage suppression. The shift to defined-contribution retirement increased households’ exposure to financial-market volatility, which made employment instability more costly and made the high-deductible health plan a more dangerous instrument. The asset conversion of housing made geographic mobility more expensive, which made the household less able to leave a job that was extracting more than it was paying. Each change made the others more consequential.
The beneficiaries of the forty-year decoupling are visible in the same federal data the decoupling is documented in. The wealthiest one percent of U.S. households held roughly thirty percent of national household wealth in 2024, up from approximately twenty-three percent in 1989. The wealthiest ten percent held roughly seventy percent. The share of national income flowing to capital rather than to labor has risen by several percentage points since the early 1970s, with the bulk of that shift accruing to the top of the income distribution. The S&P 500’s compound annual return from 1980 through 2024 was approximately eleven percent. The compound annual real wage growth for the median worker over a comparable period was a small fraction of one percent. Asset-holding has been the dominant route to economic gain in the U.S. economy for forty years, and asset-holding is concentrated.
The burden falls in the obvious places. It falls on the roughly one hundred million Americans with medical debt; on the approximately twenty-two million renter households who are cost-burdened; on the approximately half of households at risk in retirement; on the approximately one third who cannot cover a four-hundred-dollar emergency; on the roughly fifth-to-third of the workforce subject to unstable scheduling; on the workers whose real wages have lost ground over four decades. It falls on the parents who report leaving the workforce because childcare is unaffordable; on the adult children providing eldercare to parents because the institutional care system is unaffordable or unavailable; on the workers postponing medical care because of deductible exposure; on the households that cannot save because every dollar of nominal raise has been absorbed by housing, healthcare, and food cost increases that the headline inflation indicators have moved past.
And the burden falls, again, on the political imagination. A generation of Americans has come of age inside an economic discourse that describes the country as a strong economy while their personal experience is one of relentless pressure, declining buffers, and increasing exposure to shock. The dissonance is not a political problem to be managed by better messaging. It is the predictable consequence of forty years of institutional change that has decoupled aggregate growth from household security, accompanied by a coverage architecture that has not learned to talk about the gap.
The phrase doing the most work is the economy is strong — variants include GDP grew, productivity is up, the labor market is tight, wages are rising, inflation is over — used in political speech, in central bank communications, in financial press headlines, and in the political reporting that translates financial press headlines into general-interest coverage. Each of these phrases is, on a narrowly defined indicator, true. Aggregate GDP did grow. Productivity is up. The unemployment rate is low. Nominal wages have risen. Headline inflation has decelerated. The phrases are accurate within their stated boundary. The boundary is the work the phrases do.
This is narrative laundering in operation: the conversion of accurate-within-boundary indicators into a description of household economic life that the indicators were not built to provide. The discourse describes growth and asks the public to receive it as security, treating any reported gap as a failure of voter or consumer perception. The framework is reinforced by legitimacy shielding every time a household’s lived report of economic stress is dismissed as anecdotal, partisan, or unrepresentative. The asymmetry of authority between the headline indicator and the household report is itself a structural feature of the discourse. Indicators are taken seriously by default. Households are taken seriously only when their reports can be aggregated into an indicator the discourse already recognizes.
The counter-mechanisms the dominant frame rules out are well-established categories of public economic action. Sectoral bargaining would address the wage-suppression and time-theft mechanisms by establishing wage and hour floors that apply across an industry rather than firm by firm. Universal basic services — universal healthcare, universal childcare, universal eldercare, universal public education through college, public transit, public housing at scale — would address the household-absorption mechanism by removing the cost of essential services from the household balance sheet. Just-cause employment and predictable scheduling standards would address time-theft and workforce-disposability. Public risk pools would address risk-dumping. Anti-monopsony policy and non-compete bans would address the labor-market concentration that suppresses wages. Debt cancellation in medical, student, and housing categories would address the accumulated burden of the past four decades. None of these counters is hypothetical. Each is operational somewhere, often in countries with comparable per-capita GDP to the United States.
The first precedent is foreign and recent. Australia’s modern award system, codified in the Fair Work Act of 2009 and operating continuously since, sets minimum wage and working condition standards on an industry-by-industry basis through a public tribunal — the Fair Work Commission — rather than firm by firm. Approximately two and a half million Australian workers, in industries from retail to nursing to fast food to hairdressing, are covered by awards that establish minimum hourly rates above the national minimum wage, regulate scheduling, set overtime and penalty rates for evening and weekend work, and establish rest period requirements. The system has operated under both Labor and Liberal-National governments. Australia’s federal minimum wage as of 2024 was approximately AU$24.10 per hour — over three times the U.S. federal minimum at exchange-rate parity — and the modern awards floor in many industries is substantially higher. The institutional architecture is not exotic. It is one of several functional sectoral-bargaining systems in OECD economies; Germany, the Nordic countries, and France each operate variants. The U.S. has chosen not to.
The second precedent is foreign and continuous. Germany’s Mitbestimmung (co-determination) system requires that companies with more than two thousand employees include worker representatives on their supervisory boards in equal numbers to shareholder representatives. The system has operated since the Co-determination Act of 1976, with antecedents in coal and steel since 1951. It applies to several thousand of the largest German firms. Workers in those firms participate in strategic decisions about plant operations, major investment, and corporate governance through a structural mechanism rather than through bargaining alone. The German economy has produced higher productivity growth, lower wage inequality, and higher manufacturing wages than the U.S. economy over most of the past forty years, while operating under the institutional constraint that critics of co-determination predicted would prevent these outcomes. The institution has been contested — Mitbestimmung has been weakened in some sectors and through some legal challenges — but it remains operational, and the comparison between German and U.S. wage outcomes in comparable industries (auto manufacturing being the canonical case) is decisive on the question of whether co-determination produces the harms its U.S. opponents predict.
The third precedent is foreign, recent, and instructive on the architecture of universal public services. Quebec’s low-fee childcare program, established in 1997 at $5 per day and currently set at approximately CA$10 per day for the public network, provides regulated childcare to families regardless of income at a per-day cost that is a small fraction of unsubsidized U.S. childcare cost in comparable cities. The program is publicly funded, publicly administered through a network of nonprofit centers and licensed family providers, and integrated with the provincial labor force participation strategy. Quebec’s female labor force participation rate rose substantially after the program’s introduction; economists have documented that the program has substantially paid for itself through increased tax revenue from working mothers. The institutional architecture has been contested — wait lists, quality variation, and recent voucher expansions to private operators have produced mixed pressure — but the core operational record is unambiguous. A Canadian province with a per-capita GDP comparable to several U.S. states built a universal childcare access institution. The U.S. has not.
The fourth precedent is American and historical. The 1944 G.I. Bill provided returning veterans with college tuition payments, vocational training, low-interest mortgages, unemployment compensation, and small business loans, at a scale that transformed American economic life within a generation. By 1956, approximately half of all returning World War II veterans had used some component of the G.I. Bill, including roughly 7.8 million who received educational benefits and approximately 2.4 million who used the home loan guaranty. The program’s economic return — in tax revenue from higher-earning veterans, in housing construction, in higher education capacity — has been estimated at multiples of its cost. The G.I. Bill is the most consequential single piece of household-security legislation in twentieth-century U.S. history. The program was built. The country knows how. The choice not to build similar architecture for non-veteran populations is institutional, not technical.
The reframing is this: aggregate growth that does not produce household security is not what the public means by prosperity. The U.S. economic discourse has spent forty years describing growth as if its translation into security were automatic. The translation is not automatic. It is institutional. The institutions that historically performed the translation — strong unions, defined-benefit pensions, low-deductible health insurance, public investment in housing and education, predictable working time, a federal minimum wage that tracked productivity — have been systematically weakened, restructured, or never built. The growth has continued. The translation has not.
The public is not wrong about its own life. The thirty-seven percent who cannot cover a four-hundred-dollar emergency are not misinformed about their household balance sheets. The one hundred million with medical debt are not confused about whether they owe the money. The twenty-two million cost-burdened renter households are not under-counting their rent. The fifty percent at retirement risk are not over-estimating the inadequacy of their savings. The economic experience the headline indicators describe and the economic experience households report are not the same economy. The first is the aggregate of paid market activity. The second is the texture of household life within an institutional arrangement that has, for forty years, channeled growth toward asset-holders and away from wage-earners.
The illusion of prosperity is not a failure of perception. It is the structural consequence of measuring prosperity in indicators built for a different question. Aggregate output growth is a real phenomenon. So is household stress. The two have decoupled. Naming the decoupling — and naming the institutional choices that produced it — is the first analytic step in any honest U.S. economic conversation. It is also the step that mainstream economic coverage has, for forty years, not taken.
Leave a comment