Originally researched and prepared as part of public presentations with the Racial Equity Institute
Sixty years after America declared a war on poverty, a little-appreciated measure helps explain why and where poverty persists in the country today
In January 1964, in his first State of the Union address, President Lyndon Johnson declared “unconditional war on poverty in America.” By the government’s own yardstick, much of it was won. The official poverty rate has fallen from about 19% in 1964 to a near record low of 10.6% in 2024. The safety net built in those years does even more than that headline number shows: Social Security alone now lifts more than 27 million Americans out of poverty every year, and poverty among people over 65 — close to 30% in the mid-1960s — has dropped to under 10%, a decline economists credit largely to that one program.
“Unfortunately, many Americans live on the outskirts of hope—some because of their poverty, and some because of their color, and all too many because of both. Our task is to help replace their despair with opportunity. This administration today, here and now, declares unconditional war on poverty in America. I urge this Congress and all Americans to join with me in that effort.”
— President Lyndon B. Johnson, January 8, 1964
And yet the war never quite ended. In absolute numbers, almost exactly as many Americans are poor today as when Johnson spoke: 36.1 million in 1964, and 35.9 million in 2024. The rate fell chiefly because the country added roughly 150 million residents over those six decades, not because poverty was defeated. So the more revealing questions are no longer only how many Americans are poor but why poverty has proved so persistent and where it has dug in.
But our first question has to be: what does it mean to be “poor”? While the experience will always be relative, the official count is drawn with a sharp line. A person is counted as poor if their family’s pretax cash income falls below a dollar figure called a “poverty threshold,” set for that family’s size and makeup. In 2024, that line was about $16,320 for a single adult under 65 and roughly $32,130 for a family of four; the thresholds scale with family size and are adjusted each year for inflation.
Two deliberate omissions are worth keeping in mind. The measure counts income, not wealth. So, a family that owns a home, land, or other assets but brings in little cash can still fall below the line (“land rich, cash poor,” as the saying goes in farm country), while a household with savings but a low-earning year can register as poor. Because it counts only cash, it ignores non-cash help such as food assistance and housing subsidies, which is part of why the Census Bureau publishes a second, broader gauge called the Supplemental Poverty Measure alongside the official one.
Poverty is usually measured as a count: how many people fall below the federal poverty line, and what share of the population they represent. But that number says nothing about where poor families live: whether they are scattered across neighborhoods of every income level, or clustered together in places cut off from good schools, steady jobs, and basic services.
That distinction is what researchers call concentrated poverty: the share of poor people who live in high-poverty neighborhoods. In the analysis below, a “high-poverty” census tract is one where at least 25% of residents are poor (the same threshold the federal government uses to designate a HUD Qualified Census Tract).
When poverty concentrates this way, its effects compound. As Professor of Public Policy Paul Jargowsky put it in his 2013 report for The Century Foundation, a more concentrated residential pattern means more poor children grow up going to schools that function far below middle-class ones in homes with parents and caregivers with worse access to jobs and worse health outcomes and in communities with stagnant or falling property values, disadvantages that can persist across generations.
Jargowsky documented a troubling resurgence: after declining through the booming 1990s, concentrated poverty surged again after 2000, eventually eclipsing its previous 1990 peak. The maps below, built from the latest American Community Survey five-year estimates, update that picture group by group. What emerges is that the broad geography Jargowsky described is still with us and that the sharpest line running through it is still race, as evidenced in the maps that follow.
A note on comparing the two studies: Jargowsky defined high-poverty neighborhoods using a 40% poverty threshold; the maps here use the more inclusive 25% (HUD) threshold. That means the exact percentages are not directly comparable between the two, but the patterns, the rankings of states, and the contrasts between groups are. Where this new analysis compares the two, it compares geography and rank order, not raw levels.
Poverty doesn’t spread evenly. It pools.
The national map of concentrated poverty is not a smooth gradient. It has hot spots. Across all racial groups combined, concentrations are highest in Puerto Rico, Michigan, Wisconsin, Mississippi, and Louisiana, a set of places that reflects three American geographies at once: the industrial Midwest, the Deep South, and the colonial Caribbean.
This resembles the geography Jargowsky described more than a decade ago. The specific tracts have shifted, and a more inclusive threshold lights up more of the map, but the nexus of the problem has not relocated. The first thing the new maps tells us is therefore a statement about continuity: the places where American poverty persists today are, to a striking degree, the same places it persisted a decade ago.
Concentration is not a simple function of how many people are poor. It is a function of segregation and place, meaning which communities have been historically confined to which neighborhoods. For example, White Americans and Asian Americans have had similar poverty rates for the past several years, both ranging from 9% to 10%. But the two groups experience concentrated poverty in a different number of states, two states for the former and five for the latter, and in very different regions. In other words, two groups can be equally poor and live in entirely different Americas.
The map above is fine-tuned to individual census tracts to show where concentrated poverty reaches its highest levels and the areas where it clusters. But it’s also helpful to look at the broader picture. Below is another view of the data above zoomed out to the state level. The subsequent maps follow this approach for ease of viewing.
That distinction is what researchers call concentrated poverty: the share of poor people who live in high-poverty neighborhoods. In the analysis below, a “high-poverty” census tract is one where at least 25% of residents are poor (the same threshold the federal government uses to designate a HUD Qualified Census Tract).
When poverty concentrates this way, its effects compound. As Professor of Public Policy Paul Jargowsky put it in his 2013 report for The Century Foundation, a more concentrated residential pattern means more poor children grow up going to schools that function far below middle-class ones in homes with parents and caregivers with worse access to jobs and worse health outcomes and in communities with stagnant or falling property values, disadvantages that can persist across generations.