The American Dream Affordability Index: A Composite for 2026

What an affordability index has to do

Headline inflation numbers from the Bureau of Labor Statistics Consumer Price Index describe the average price change of a national basket of goods. They do not describe whether a working household in Boise or Charlotte or Phoenix can still buy the bundle that used to define "middle class" in the United States: a modest house, two reliable cars, employer-linked health insurance, childcare for two children, and a state-school tuition pathway.

An American Dream affordability index has to translate that bundle into a price, then divide by the median income of households who would plausibly purchase it, then track the ratio across metro areas and over time. The math is not exotic. The data discipline is.

The five-component bundle

There is no perfect bundle, but the five categories below capture roughly 80 percent of the variance in cost-of-living differences across US metro areas based on the BLS Consumer Expenditure Survey.

• Housing. Either the median monthly mortgage payment on a median-priced home with a 10 percent down payment at the prevailing 30-year rate, or the HUD Fair Market Rent for a two-bedroom unit, whichever the household type uses. Zillow's median home value index and Redfin's housing market data are the standard sources.
• Childcare. Child Care Aware of America publishes state-level annual cost averages. For two children, the national average exceeded $24,000 in 2023, more than median rent in most metros.
• Health insurance. Family-coverage premiums from the Kaiser Family Foundation Employer Health Benefits Survey plus average deductible exposure. For households without employer coverage, the marketplace benchmark premium net of subsidies.
• Transportation. Either the AAA Your Driving Costs annual estimate for one or two vehicles, or transit pass cost where transit can substitute. The BLS Consumer Expenditure Survey is the cross-check.
• Higher education. A four-year cost-of-attendance figure for the flagship state university, prorated over the years a household with school-age children is saving toward it. The College Board Trends in College Pricing report is the standard reference.

Sum the five, compare to the local median household income from the American Community Survey, and the ratio is the index. A ratio under 0.7 historically corresponds to comfortable middle-class affordability. A ratio at or above 1.0 means the bundle costs more than the median household earns.

What the ratio looks like in 2026

The Joint Center for Housing Studies at Harvard documents in its annual State of the Nation's Housing report that the share of cost-burdened households (paying more than 30 percent of income on housing) reached 22.4 million in 2023, an all-time high. Combine that with childcare costs that grew faster than median wages in every state between 2018 and 2024, family health insurance premiums that rose 24 percent over the same period per the KFF survey, and the affordability ratio is over 1.0 in most major coastal and Mountain West metros for households earning the local median income.

The Brookings Hamilton Project has tracked this composite under the label middle-class index and reaches similar conclusions: the metros where the index has deteriorated fastest since 2010 are Phoenix, Boise, Nashville, Charlotte, and Tampa, not the coastal metros that dominated the previous decade's headlines.

Why AI shows up in the affordability conversation

Two channels matter.

First, AI changes the wage distribution inside exposed occupations. If the productivity gain accrues to a small share of the workforce while displacing others into lower-paying roles, median household income, the denominator of the affordability ratio, can stagnate even as national output grows. This is the substitution-then-reabsorption story studied by Acemoglu, Autor, and others, summarised in Autor's 2024 Nobel-cited essay and in our county-level companion piece, AI job displacement by county.

Second, AI changes the cost of the bundle. AI-driven productivity in residential construction, in primary care, in childcare administration, and in higher education delivery could lower the numerator. So far the empirical case for cost reductions on big-ticket household categories is weaker than the case for displacement on the wage side. The Joint Economic Committee's 2024 issue brief on AI productivity argues this gap will close; the OECD review is more cautious.

The net effect on the affordability index depends on which channel moves first. A future in which AI improves wages without lowering bundle costs makes the ratio worse in dollar terms but does not necessarily change the ratio. A future in which AI lowers bundle costs without raising median wages improves the ratio for everyone. The political economy of the next decade is about which path the country actually takes.

How to build an honest index

Any composite has to publish four design choices on the methodology page.

• The choice of household structure, because a single-adult, no-children household and a two-adult, two-children household have very different bundles. Most indices publish the four-person household as the headline and the single-adult version as a sensitivity check.
• The treatment of regional price parities. The Bureau of Economic Analysis regional price parities adjust for local cost differences but exclude housing if the index already uses HUD or Zillow data, to avoid double-counting.
• The income denominator. Median household income from ACS is the standard, but it conflates households at different life stages. Reporting separate ratios for working-age households (25-54) and for households with school-age children sharpens the picture.
• The vintage. Housing data is monthly, KFF premiums are annual, ACS median income is annual with a roughly 18-month lag. Mixing vintages without disclosure is the most common error in casual affordability rankings.

The point is not to discourage composites. It is to insist that any index whose conclusions land in a city council meeting or a presidential debate has to be reproducible from public data.

What states and cities can do

The local levers are not mysterious: zoning reform that allows additional housing supply at the price points the bundle actually requires, expanded eligibility and supply for childcare subsidies, transit investment that lets households substitute away from a second vehicle, and tuition-stability commitments at state flagships. The American Enterprise Institute housing center and the Urban Institute publish parallel research on which of these levers move the index fastest. Our piece on Geography of Opportunity Rankings covers the structural mobility differences that interact with affordability over a child's lifetime.

The takeaway is that the American Dream is still measurable, but only if you are willing to measure the bundle, not the slogan. An index that does so is a planning tool. An index that does not is a press release.

Sources

• US Bureau of Labor Statistics, *Consumer Price Index* and *Consumer Expenditure Survey*.
• Joint Center for Housing Studies of Harvard University, *State of the Nation's Housing 2024*.
• Kaiser Family Foundation, *2024 Employer Health Benefits Survey*.
• Child Care Aware of America, *Price of Care: Annual Report*.
• College Board, *Trends in College Pricing*.
• US Census Bureau, *American Community Survey*.
• Bureau of Economic Analysis, *Regional Price Parities*.
• Brookings Hamilton Project, *Vitality Index and Middle-Class Indicators*.