Which states had the highest estimated share of people in poverty in 2025? Louisiana and Mississippi were among the highest, while New Hampshire had the lowest state estimate in the U.S. Census Bureau’s American Community Survey. The comparison of poverty rates across states uses one survey and one measure; the small difference between the two highest estimates is not statistically significant.
Highest poverty rate in US survey estimates
The highest poverty rate in US state estimates was 18.1% for Louisiana, followed closely by Mississippi at 18.0%, according to the Census Bureau’s 2025 state poverty report. Louisiana is the state with the highest poverty rate by the point estimate among the 50 states, but the bureau says the two estimates are not statistically different. That means the data do not establish a reliable first-place distinction between them.
New Hampshire’s estimate was 6.2%, the lowest among the states. The table below gives selected examples at both ends, using the same Census report and the 2025 American Community Survey (ACS) one-year estimates. The margins of error are in percentage points and use a 90% confidence level.
| State | Estimated share below poverty threshold | Margin of error |
|---|---|---|
| Louisiana | 18.1% | ±0.8 points |
| Mississippi | 18.0% | ±0.7 points |
| West Virginia | 16.0% | ±0.8 points |
| New Hampshire | 6.2% | ±0.6 points |
| Utah | 8.6% | ±0.6 points |
| New Jersey | 8.7% | ±0.4 points |
These are selected examples, not a complete ranking. They show why state poverty rates should be read alongside uncertainty: a displayed difference of a tenth of a percentage point is much smaller than the individual margins shown for Louisiana and Mississippi. The Census Bureau’s statistical test, rather than a comparison of rounded figures alone, supports its conclusion about those two estimates.
What the percentages measure
The ACS poverty rate by state is the estimated share of people in the survey’s poverty universe whose family or individual income fell below the applicable official poverty threshold. The thresholds vary with family size and composition. The bureau compares them with money income before taxes, including earnings and certain cash benefits.
The calculation does not include noncash assistance such as nutrition benefits or housing subsidies, and it does not count tax credits as income. The thresholds are updated for inflation but do not vary by state to reflect local housing costs. As a result, these figures answer a specific income-based question; they do not describe every household’s expenses or access to assistance.
Poverty rates by state also omit some populations from the calculation. The Census Bureau excludes, among others, people living in institutional group quarters, college dormitories, and military barracks. The denominator is therefore not identical to every person physically present in a state. The report describes the full poverty universe in its methodology notes.
Understanding margins of error
The ACS surveys a sample rather than interviewing every resident. A margin of error expresses sampling uncertainty around an estimate. For example, New Hampshire’s 6.2% estimate with a ±0.6-point margin produces an interval from 5.6% to 6.8% at the 90% confidence level. A 90% confidence procedure is designed to contain the true population value in about 90% of intervals produced by repeated samples using the same method.
The uncertainty is a reason to avoid treating every decimal as a precise position. Overlapping margins alone do not provide a definitive significance test: use the Census Bureau’s published comparison or an appropriate statistical test to assess a difference. Estimates can also be affected by nonsampling issues, such as incomplete responses or measurement error, which the reported sampling margin does not fully capture.
Comparing rates, counts, and places
A percentage describes a share of the population measured; a count describes how many people are estimated to fall below the threshold. These answer different questions. A large state can have a lower percentage than a smaller state and still have more people below the threshold because its population is larger. When someone asks about the states with most poverty, clarify whether they mean the number of people or the proportion of residents.
The states with highest poverty estimates in the table are identified by their shares, not by a count of people. Within each state, conditions also vary across counties, neighborhoods, age groups, and families. A state average is useful for a broad comparison, but local data can be more informative for a community organization deciding where and how to offer services.
The table compares the 50 states. The Census report also publishes estimates for the District of Columbia and Puerto Rico, which are separate jurisdictions rather than states. Include them deliberately when using the report’s full geographic table, and state the comparison group so readers can understand what a ranking covers.
Reading other poverty measures
Published poverty levels by state may differ because a source uses another survey, time period, population, or definition. The Supplemental Poverty Measure, for example, accounts for some benefits and expenses and adjusts for housing costs in ways the official measure does not. Its results answer a related but different question, so they should not be mixed into the same list as these ACS official-measure estimates.
When comparing a new source with this table, check the year, the measure, the survey, the geographic coverage, and the margin of error first. For a change over time, use comparable estimates and consult the reported test of the change; an increase or decrease in rounded numbers alone may not be statistically meaningful. For smaller places, a multi-year estimate may be available and should be labeled with its full period rather than presented as a single year.
