There is an ongoing and lively debate over the size and composition of the deep poor population (commonly defined as those living below half the poverty threshold), how deep poverty should be measured, and whether the percentage of the United States population living in deep poverty has increased or decreased over the past 20 years. Nonetheless, evidence suggests that a portion of the U.S. population lives on very little cash income. Some analysts have argued that welfare policy shifts in the 1990s, which reallocated public benefits from nonworking to working households and increasingly prioritized in-kind benefits over direct cash transfers, are important factors. Whereas low-income working families benefited from these policy changes, families disconnected from the formal labor market appear to have fared less well as eligibility for public assistance became increasingly contingent on work. At greatest risk of deep poverty are families headed by a single parent, formerly incarcerated individuals and their children, racial/ethnic minority populations, immigrants, youths aging out of foster care, and individuals who are not working or receiving cash welfare. For many of these families, instability in employment and inadequacy of hours and wages, rather than not working at all, appears to be predominant (although long-term unemployment is also a significant factor). Moreover, many such families face barriers to sustained employment such as physical and mental illness, disability, addiction, and lack of transportation. This brief explores current understanding of deep poverty and identifies important areas for new research to better inform policy aimed at preventing and reducing deep poverty and its effects.