Help Me--and the U.S.--Define Poverty
Updated: Aug 17
We know our Official Poverty Measure (OPM) is wrong, but we still use it to redistribute $2.9T.
Mollie Orshansky never meant to define poverty. As a statistician for the Social Security Administration, she just wanted to know which families with children couldn't afford a basic diet. So in 1963 she derived a minimum income threshold using the USDA's cheapest food plan, multiplied by three to account for other expenses.
She never meant for it to become law. But a year and a half later, the Office of Economic Opportunity — the agency running Johnson's brand-new War on Poverty — needed a number, and Orshansky's happened to be sitting right there. They adopted it enthusiastically, one official even calling it a "second generation definition of poverty." A researcher's back-of-envelope estimate, built to flag which families needed help most, calcified into the permanent, official yardstick for an entire country. Orshansky spent the rest of her career objecting to exactly that — she built a tool, and the government misappropriated it into a monument.
Sixty years later, we're still using her monument. It doesn't account for rent, child care, education, or literally anything other than the 1963 USDA food plan, adjusted for inflation. Worse, the "income" we measure against it is pre-tax earned income — disregarding the roughly $1.9 trillion a year we spend on transfer payments for people in need. Count that money as what it is — income — and the official 11% poverty rate falls to somewhere between 2.5% and 4%, depending on whose methodology you trust. Different studies land in different spots, but they all agree on the direction: official poverty is a wildly overstated number.
That's still not the full story, though. We need a more comprehensive definition of poverty.
We need to add up all resources — earnings, assets, and benefits — subtract the costs of life, then somewhere along that continuum draw a line in the sand and decree, "Anyone below this line is poor." As you can imagine, there are thousands of ways to do this, which makes it stranger that we don't have one official way of identifying the most destitute among us.
A better guess is our Supplemental Poverty Measure (SPM). This time we include post-tax dollars and benefits, then subtract most of life's real expenses — work costs, child care, taxes, and medical spending out of pocket. This net resources measure shows what's actually left after life takes its cut. The threshold here is set at roughly 83% of the median household's spending on food, clothing, shelter, and utilities. Sounds a lot better! But wait. Poverty is as much about the "who" as it is about "how many". "Who" is what the SPM gets wrong.
According to the SPM, wealthy old women are poor. Turns out older women spend a lot maintaining their households and appearances — so much so that once you subtract those costs from their income, they look poor on paper. But older people live more off passive income, like investments and Social Security, than they do off a paycheck. This is one example of how being income poor doesn't mean you're consumption poor.
Consumption is a better indicator of poverty. How much we consume is already the core of the utility functions economists use to determine well-being. We also lean on savings, credit, and benefits to maintain our lifestyle when paychecks are cutting it close. Thankfully, the BLS conducts quarterly surveys at the household — or "consumer unit" (CU) — level. The Consumer Expenditure Survey is generally regarded as the best data we have on the income, consumption, and federal aid flowing through American homes.
Below, I looked at households at the same level of consumption, opened the door, and peeked inside at our many American lives. As we move down the consumption scale, Americans are more likely to be people of color, single parents, renters, and non-college graduates — and much more likely to be receiving benefits.
Rather than debating "what is poverty?" in the abstract, shift the consumption threshold yourself and watch how your definition changes who counts as "poor." Click the menu in the upper left corner, set the poverty threshold relative to median consumption, and watch the picture of poverty — who, not just how many — evolve in front of you.
