Fast Focus Research/Policy Brief Icon

U.S. Federal and State Poverty-Reduction Policy: Divergence in Discretion

  • January 2024
  • Fast Focus Policy Brief No. 63-2023

Poverty-reduction policies in the United States are inconsistent in reducing poverty for households with children compared to those without children.

State and federal policies designed to reduce or eliminate poverty tend to divide populations into categories to determine eligibility for financial assistance. These “categorical eligibility structures” sort people and households into groups—such as children, workers, or senior citizens—who may be eligible for certain safety-net benefits based on income, age, location, or other demographic characteristics. Anti-poverty policies also delegate elements of program design and administration to states and localities, where variations can be substantial and opportunities for discrimination are multiplied. Fragmenting populations into categories undermines the universal application of social policy while jurisdictional fragmentation undermines policy’s capacity for equity.

Recent research by Sarah Bruch, Joseph van der Naald, and Janet Gornick examines how the effects of state and federal social welfare policies are distributed throughout the United States and across demographic categories of people.[1] Bruch and colleagues evaluated U.S. Census Bureau Current Population Survey data from 1996 to 2016 to assess the effectiveness of policy mechanisms—asking the basic question: “Did this policy reduce poverty?”—in working-age households both with and without children. In short, Bruch and colleagues find the patchwork assortment of social welfare and anti-poverty programs in the United States exhibiting significant state-by-state variation in actual poverty reduction.

Cash (and near-cash) transfers and taxation are two primary mechanisms for implementing social welfare policy. These occur at both the federal and state levels. Policy at the federal level—treating citizens of all states as equal—is an example of centralized policy implementation. Policy generated and implemented at the state level—where variations can be substantial—exemplifies decentralized policy implementation. Two important consequences of a decentralized social safety net include: inconsistent promotion of economic well-being across household types (e.g., those with and without children) and geographic inequalities state-by-state.

The patchwork assortment of social welfare and anti-poverty programs in the United States exhibits significant state-by-state variation in actual poverty reduction.

Eligibility, Deservingness, and Economic Well-Being

Qualifying for assistance when facing economic hardship requires meeting eligibility criteria often based on the demographic characteristics of an individual or household. For example, senior citizens and workers tend to qualify for federal sources of assistance through social insurance policies based on contributions by employers and employees over the course of their work life. Parents of children under 18 years-old are eligible for programs generally based on income level (i.e., means-tested) and/or participation in the paid labor force. These programs tend to vary in terms of overall quality and quantity of benefits, and in the policy mechanisms at play (e.g., cash transfer, in-kind, service-based, or tax expenditure), as a mix of centralized and decentralized sources.

Persistent, often negative, narratives about people experiencing poverty in the United States and elsewhere have developed and grown through the “War on Poverty” era of the 1960s to today. Often these narratives relate to ideas of deservingness. Such narratives can also amplify other persistent biases related to issues of race, class, and gender. As a result, program recipients are easily stereotyped as more deserving or less deserving of public assistance based on perceived demographic differences. Children, for example, have been consistently seen as a group deserving of assistance and are the topic of much poverty-policy research. One of the most consistent findings of this research is the predominant role of federal income tax credits (e.g., the Earned Income Tax Credit, or EITC) and food assistance programs (e.g., Supplemental Nutrition Assistance Program, or SNAP) in buffering against child poverty over the past two decades.[2]

Persistent, often negative, narratives about people experiencing poverty in the United States and elsewhere have developed and grown through the “War on Poverty” era of the 1960s to today.

Between 1996 and 2016, economic well-being generally increased for many households facing economic hardship in the United States. This reduction in poverty was significantly larger, on average, for working-age households with children compared to households without children. A central finding for Bruch, van der Naald, and Gornick involved larger poverty reductions (i.e., more effective policy) for working-age households with children than for those households facing economic hardship without children; these larger reductions also generally increased across the years 1996 to 2016. A consequence of this disparity in poverty reduction is a growing gap in rates of disposable-income poverty between households with kids and those without; as designed, the social safety net appears more responsive to increased market-income poverty for households with children.[3]

Market Income
Pre-tax, pre-transfer income
Includes: Wages + salary, self-employment earnings, farm income, retirement, survivor pensions, disability pensions, and annuities, etc.
Disposable Income
Post-tax, post-transfer income
Includes: All market-income sources as well as federal transfers, state transfers, federal taxes, and state taxes, etc.

Implications of Taxes and Transfers as Policy Mechanisms

The design of poverty-reduction policies in the United States—including the key mechanisms of categorical eligibility criteria, income targeting, and decentralized administration—are not consistently effective in promoting economic well-being (i.e., reducing poverty) for households with children compared to those without children. Bruch, van der Naald, and Gornick distinguish between two primary policy mechanisms—taxes and transfers—at both the federal and state levels. These four mechanisms—federal taxes, federal transfers, state taxes, and state transfers—examined across 20 years and all 50 states produce several important insights.

For households whose incomes falls below the poverty line, household structure matters a great deal for the level of assistance received.[4] Household structure is important in two primary ways. First, available programs differ considerably for households with children and for parents able to work in the paid labor force. Second, assistance via tax credits varies significantly depending on the number of dependents in a household. At the same time, the amount of assistance households with children can receive is heavily influenced by state-level policymaking that, even within this household type, has tremendous variation due to the political geography of the United States. Moreover, these dynamics shift slightly in response to national economic fluctuations (sometimes called “cyclical responsiveness”). Taken together, by demonstrating how poverty reduction shifts over time and across states, Bruch and colleagues highlight the negative impacts of U.S. safety net fragmentation.
Medium #c8dbe0
Low #e8ebd1

Table 1: Categories of State Transfer Programs by Levels of State Discretion (Low, Medium, High)

Financing Rulemaking Administration
General assistance High High High
Workers’ compensation High High High
Temporary Assistance for Needy Families High High High
Child support Medium Medium High
Unemployment insurance Medium Medium Medium
Low-income Home Energy Assistance Program Low Medium High
Public housing and rental subsidies Low Medium High
Supplemental Nutrition Assistance Program Low Low Medium
National School Lunch Program Low Low Medium
Supplemental Security Income Low Low Low
Note: Low = limited state discretion; High = a great deal of state discretion. Based on program design features distributing federal, state, and local government responsibilities and authority; coding by Bruch, van der Naald, and Gornick (2023).

Another primary finding demonstrates how low market-income households with children have access to a wider array of assistance; in these households, poverty is reduced to various extents by all four policy mechanisms (i.e., federal and state taxes and transfers). In contrast, households without children who also grapple with low market income can access assistance through federal and state transfers, yet federal and state taxes tend to push these households further into poverty. In long-standing arguments about the social construction of deservingness, children are seen as a particularly deserving population for whom numerous safety net programs are designed to serve. Working-age households without children, however, are seen as generally less deserving unless an extenuating social status (e.g., veterans, people with disabilities) categorizes them into eligibility for assistance.

States with higher percentages of Black citizens tend to have more oppressive, paternalistic, and punitive policy designs for safety net implementation.

State-level decisions regarding transfer programs can vary widely in terms of financing, administration, or rulemaking. Regardless of whether a program is considered low, moderate, or highly discretionary at the state-level (see Table 1), variations result in geographically bounded differences in quantities, qualities, and access to safety net assistance across states. Many programs intended to assist households with children function in a relatively decentralized manner. Unemployment Insurance (UI), the Supplemental Nutrition Assistance Program (SNAP—formerly known as “food stamps”), Supplemental Security Income (SSI), and Temporary Assistance for Needy Families (TANF), worker’s compensation, and school meal subsidies are all decentralized transfer programs. It is notable that the two most influential state transfer programs—SNAP and SSI, which contribute the most to poverty reduction for households with children—allow relatively low discretion by states.

Inconsistent application of program breadth and effectiveness is particularly troubling when consistent differences emerge along racial lines. States with higher percentages of Black citizens tend to have more oppressive, paternalistic, and punitive policy designs for safety net implementation.[5] These states also tend to be less inclusive and less effective in reducing poverty for racialized populations.[6] Decentralized “discretion by design”—a key distinction between state and federal safety net programs—is a mechanism of policy implementation that reinforces systemic inequalities in the United States. The role of decentralization in this policy context is a topic ripe for further research by scholars interested in poverty-reduction policy and its effective implementation nationwide.

Compiled and edited by James T. Spartz, with editorial and design support from Judith Siers-Poisson, Nateya Taylor, and Dawn Duren.

[1]Bruch, S. K., van der Naald, J., & Gornick, J. C. (2023, June). Poverty reduction through federal and state policy mechanisms: Variation over time and across the United States. Social Services Review, 97(2), 270–319.

[2]Bitler, M., Hoynes, H. & Kuka, E. (2017). Child poverty, the Great Recession, and the social safety net in the United States. Journal of Policy Analysis and Management, 36(2), 358–389.; Pac, J., Nam, J., Waldfogel, J., & Wimer, C. (2017, March). Young child poverty in the United States: Analyzing trends in poverty and the role of anti-poverty programs using the Supplemental Poverty Measure. Children and Youth Services Review, 74, 35–49.; Pac, J., Waldfogel, J., & Wimer, C. (2017). Poverty among foster children: Estimates using the Supplemental Poverty Measure. Social Service Review, 91(1), 8–40.

[3]Brady, D. & Parolin, Z. (2020). The levels and trends in deep and extreme poverty in the United States, 1993–2016. Demography, 57(6), 2337–23360.; Wimer, C., Parolin, Z., Fenton, A., Fox, L., & Jencks, C. (2020). The direct effect of taxes and transfers on changes in the U.S. income distribution, 1967–2015. Demography, 57(5), 1833–1851.

[4]ASPE, Office of the Assistant Secretary for Planning and Evaluation. Poverty Guidelines.

[5]Soss, J., Fording, R. C. & Schram, S. F. (2011). Disciplining the poor: Neoliberal paternalism and the persistent power of race. Chicago: University of Chicago Press.

[6]Bruch, S. K., Gornick, J. C. & van der Naald, J. (2022). Geographic inequality in social provision: Variation across the U. S. States. In Measuring Distribution and Mobility of Income and Wealth, edited by R. Chetty, J. N. Friedman, J. C. Gornick, B. Johnson, & A. Kennickell, 499–528. Chicago: University of Chicago Press. Gaines, A. C., Hardy, B. & Schweitzer, J. (2021). How weak safety net policies exacerbate regional and racial inequality. Center for American Progress.


Child Poverty, Children, Economic Support, Economic Support General, Inequality & Mobility, Place, Place General, Racial/Ethnic Inequality


, , , , ,