The American dream got expensive because it was never a single purchase. It was a bundle: a house, a car, health coverage, a child’s education, and a retirement, historically supported by one or one and a half incomes. Each component inflated at its own rate over four decades. Wages tracked some of them and fell behind the rest. The compounding gap between the bundle’s total cost and household income is what people are describing when they say the arrangement stopped working, and it is measurable.
The bundle, priced separately
Take the components one at a time.
The house
Figures from the National Association of Realtors and the U.S. Census Bureau put the median home sale price between $400,000 and $420,000 as of 2024. The Census Bureau reported median household income near $80,000 for 2023. That puts the median home at roughly five times median household income. On Census Bureau and National Association of Realtors historical series, the same ratio ran closer to three times in the 1980s.
This ratio does more explanatory work than any other number in the discussion. A house at three times income is reachable with a normal savings rate and a normal career. A house at five times income requires either a much longer accumulation period, a larger family transfer, or a two-earner household with no interruption. The shift changed who can buy and when, not merely how much buyers pay.
Health coverage
KFF reported total family health coverage premiums near $25,000 per year as of 2024, with the worker’s share exceeding $6,000. The remainder is paid by the employer, which economists generally treat as compensation the worker does not see. Measured that way, a substantial slice of what would otherwise be wage growth was absorbed by health premium growth over the past several decades. Wages did not stagnate so much as get rerouted.
Medical debt is the visible residue. A KFF analysis of Census data, published in 2022 and reflecting 2021, found at least $220 billion in medical debt outstanding, and a KFF and NPR investigation in 2022 estimated roughly 100 million adults carrying some form of health care debt.
Child care
Child Care Aware reports center-based care commonly running $10,000 to $17,000 annually per child, and higher in some states. Bureau of Labor Statistics data show the day care and preschool index rising 28.5 percent between December 2019 and December 2025. This category barely existed as a major line item in the single-earner arrangement, because one parent typically provided the care. When the second earner became necessary to afford the house, child care became a purchased service, and its cost partially offset the second income.
Education
The Education Data Initiative puts average student loan debt near $38,000 per borrower, and the Federal Reserve G.19 consumer credit release places total outstanding student debt between $1.7 trillion and $1.77 trillion. A credential that once paid for itself within a few years of graduation now arrives attached to a payment that competes with the mortgage the credential was supposed to enable.
Everything else
The routine costs moved too. BLS data from December 2019 to December 2025 show electricity up 40.7 percent, food at home up 30.6 percent, used vehicles up 33.6 percent, and hospital services up 30.4 percent. Over the same window, average hourly earnings rose 30.4 percent and the all items consumer price index rose 26.1 percent.
Why matching averages does not restore the bundle
Wages beat overall inflation slightly across that six-year window. Anyone arguing that households therefore gained ground is making an error worth naming precisely.
The all items index measures a consumption basket. The bundle is not a consumption basket. It is a set of assets and long-lived commitments, and three of its largest components, housing, education, and health care, compounded at rates above general inflation for decades before this six-year window opened. Matching general inflation from 2019 forward leaves that accumulated divergence entirely in place.
A second issue is that the fastest-rising categories are the least substitutable. A household facing higher grocery prices can change brands. A household facing a five times income home price cannot buy four-fifths of a house. Inflation concentrated in indivisible, non-optional goods hits harder than the same average spread across discretionary ones.
Third, the bundle assumed a single income. Its modern version usually requires two, which means the household has lost the reserve it once held. When one earner covered the bundle, the second adult was insurance against job loss, illness, or a bad year. Once both incomes are committed, that insurance is gone, and the same household that looks financially adequate on paper has no capacity to absorb a shock.
What the wage debate gets right and wrong
The federal minimum wage has sat at $7.25 since 2009, per the U.S. Department of Labor. That is a real fact about a real policy failure, and it matters most for workers at the bottom of the distribution.
It is also insufficient as an explanation. The households describing the dream as out of reach are frequently earning well above any minimum wage. A household at the median income of roughly $80,000 is not constrained by the wage floor. It is constrained by a home price at five times income, a family health premium near $25,000, and child care that can run $17,000 per child. Raising the wage floor does not address any of those three.
This is the analytical core of the affordability argument, which holds that the cost side deserves at least equal attention to the wage side. Several organizations working on the issue have written at length on why the American dream stopped working as a coherent set of expectations rather than as a slogan, and the framing is more useful than treating any single price as the culprit.
What would move the ratio
The five times income figure is a fraction, and fractions move from either direction. Wage growth alone would have to be extraordinary and sustained to close it, and it would face the problem that higher incomes bid up a fixed housing stock, pushing the numerator up alongside the denominator.
The supply side is where the arithmetic is more favorable. More housing units per parcel, faster permitting, and smaller allowable lot sizes each reduce the cost of putting a household under a roof. Similar logic applies to child care capacity and to the administrative cost layered into health care delivery.
None of that is quick. Housing supply responds on a multi-year lag, and the components of the bundle are governed by different institutions at different levels of government, which is precisely why the problem has proven durable. The bundle broke one component at a time, over forty years, and it will not reassemble faster than that unless several of those institutions move at once.