A 1963 Food Budget Still Draws the Poverty Line

For all the sophistication of modern government statistics, the official measure does not count food stamps, rental subsidies, or refundable tax credits. A second yardstick, the Supplemental Poverty Measure, was built explicitly to capture what the old one misses, and the two now coexist because they answer different questions about what it costs to stay out of poverty.

The Multiplication at the Heart of the Official Line

In August 1969 the U.S. Bureau of the Budget designated the Census Bureau’s poverty thresholds as the federal government’s official statistical definition of poverty. The thresholds rested on a single calculation: “three times the cost of a minimum food diet in 1963.” Household spending surveys from the era suggested that families of three or more dedicated about a third of their after-tax income to food. So the formula took the price of a no-frills nutrition basket built by the Department of Agriculture—and tripled it. That produced the poverty line for a family of a given size and composition, and the only thing that changes year to year is the price index. The Census Bureau updates the thresholds with the Consumer Price Index for All Urban Consumers, the same CPI-U that adjusts Social Security checks, but the diet on which the arithmetic rests has never been re-costed.

The official thresholds do not vary by geography. A family of four in Manhattan faces the same dollar cutoff as an identical family in rural Mississippi, because the 1963 calculation treated the country as a single kitchen. The thresholds also vary only by family size and the age mix of members—an extra child shifts the line upward, but nothing else in the formula acknowledges that a dollar buys far less shelter in a coastal city than in the interior.

What That 1963 Formula Leaves Out

The official measure compares a family’s pretax cash income against the threshold. Wages, Social Security benefits, alimony, workers’ compensation—those all count. But enormous swaths of actual household resources do not. Nutrition assistance from SNAP, subsidized housing, the Low Income Home Energy Assistance Program, and broadband-internet assistance are invisible to the count. Refundable tax credits, particularly the Earned Income Tax Credit and the Child Tax Credit, are also ignored. A family that receives food benefits and a lump-sum credit at tax time can be cash-poor on paper while the official number says they are still in poverty.

Because the measure tallies only cash income before taxes, it cannot see the government’s own efforts to reduce hardship. A dollar of SNAP benefits does as much to put food on a table as a dollar of wages, but it never moves the dial on the official rate. Census analysts have noted for decades that this narrow resource definition misrepresents the effect of anti-poverty programs, making it hard to know whether a new tax credit or housing voucher actually lifts families above the line.

A Better Snapshot of 21st-Century Hardship

The Supplemental Poverty Measure was designed to fix those blind spots. Instead of a food-times-three threshold frozen in time, the SPM builds its line from what Americans actually spend on food, clothing, shelter, utilities, telephone, and internet—a basket the Bureau of Labor Statistics labels FCSUti. The number comes from the Consumer Expenditure Survey and therefore reflects current consumption patterns, not a diet plan from the Kennedy years. On the resource side, the SPM includes cash income plus the value of in-kind benefits, then subtracts necessary expenses: federal and state income taxes, payroll taxes, child care, work-related costs, and out-of-pocket medical care. The subtraction of medical expenses makes the SPM particularly sensitive to the health costs that often push older adults below the line.

Geographic adjustment is built into the SPM threshold. Housing costs, the largest difference in living expenses across the country, are factored in so that the poverty line rises in high-rent markets and falls where shelter is cheap. A single national number stops making sense once you accept that the same dollar buys wildly different amounts of housing in San Francisco and in the Mississippi Delta. This adjustment alone accounts for some of the largest shifts the SPM reports compared to the official numbers.

How BLS Builds a Threshold Out of Today’s Receipts

The SPM thresholds are the product of a specialized unit inside the Bureau of Labor Statistics, the Division of Price and Index Number Research. Working from Consumer Expenditure Survey microdata, the division calculates what households at the 33rd percentile of spending on the FCSUti basket pay—adding a small multiplier to cover “other necessities”—and then adjusts for geography. The thresholds are updated every year, so they move not only with inflation but also as actual spending shifts. If Americans begin spending a larger share of their budget on internet access, the threshold absorbs that change automatically.

What results is a poverty yardstick that is conceptually closer to a relative-income measure, though it remains firmly anchored to observed spending. The Census Bureau reports that the SPM and the official measure produce different rates precisely because they use different resources, different thresholds, and geographic adjustments. A child who appears poor under the old food-budget arithmetic may not look poor under the SPM, thanks to the inclusion of tax credits and nutrition aid. Meanwhile, an older person with modest income but steep medical and prescription drug bills can be recorded as poor by the SPM while the official measure ranks them above the line.

Two Numbers Every Year, and They Are Not the Same

There is one more complication worth untangling. The poverty thresholds that the Census Bureau uses for statistical tables are not the same thing as the “poverty guidelines” that the Department of Health and Human Services issues each year. HHS derives the guidelines from the Census thresholds, rounding and simplifying for administrative use. They are the numbers that determine eligibility for programs like Medicaid, SNAP, and the Children’s Health Insurance Program. For a recent year, the Census Bureau’s Small Area Income and Poverty Estimates show the official threshold for a family of four with two related children under 18 at $31,812. The HHS poverty guideline for the following year, by contrast, is $32,150 for a family of four. The small gap is a reminder that a figure that began as a statistical abstraction now runs twin tracks through federal law.

That duality—a statistical measure and an administrative guideline, an old arithmetic and a newer one—means no single number tells the whole story. The food-budget threshold persists because it offers a consistent historical series that stretches back decades, something the SPM cannot do. The SPM responds to contemporary spending, to tax policy, and to the uneven map of housing costs; it catches what the official measure erases. One measure tells you whether cash income is keeping pace with 1963-era expectations. The other tells you whether the full set of government transfers and living expenses leaves a household above or below a modern floor. Together they make visible what each one alone conceals.