Association for Public Policy Analysis & Management

Background: US social welfare policy is governed by a labyrinthine set of rules that define program eligibility, enrollment procedures and the cash value of benefits received.  The delegation of the administration of many US safety-net programs to the states has created wide variations in overall program generosity, which ultimately has a substantial impact on state poverty levels and the direct program benefits that people receive. This complex set of rules has impeded researchers’ ability to achieve an overall picture of which states have the most and least generous social-safety nets.

Although organizations have generated a number of program-specific datasets that document state program eligibility rules (e.g., the Urban Institute’s Welfare Rules database; USDA’s SNAP policy database; Kaiser Family Foundation’s state Medicaid policy spreadsheets), researchers have not developed a comprehensive dataset on safety-net rules that: 1. Combines measures from a variety of different programs into a single, decomposable index; 2. Scores programs rules in terms of their degree of generosity; 3. Simultaneously measures eligibility rules and the cash value of benefits received; 4. Examines measures over a long time period. This paper describes the methodology for constructing the index, rankings of states based on these indicators and initial predictive validity tests.

Methods: We developed a composite index of four major safety-net programs (TANF, SNAP, Medicaid, UI) with eligibility rules and benefit levels that vary by state between 1996-2012. We compiled data from a number of publicly available data sources and through primary data collection (Westlaw and the IRS TAXSIM model). States with eligibility rules that make signing up for programs easier and go beyond minimum federal requirements were scored higher whereas states doing only the minimum required by law were scored lower. We then summed points for each program and divided by the total possible points (0-1 scale). For cash benefit generosity, we modeled benefits based on a notional individual adjusting for income and FICA tax and generated income replacement rates (0-1 scale). We averaged these to get an overall (decomposable) eligibility generosity index.  To test the predictive validity of the index, we analyzed correlates of state generosity including state poverty, median household income, inequality, and racial composition.

Results: The most and least generous states varied across programs and over time. The most and least generous states were not always those predicted based on state ideology or wealth, but on average states whose citizenry were more liberal and wealthier states had more generous TANF and SNAP eligibility rules, but not more generous Medicaid eligibility rules.  States also varied in the generosity of their cash benefits- states with generous eligibility rules tended to have less generous replacement rates.

Conclusions:  States vary tremendously in overall safety-net generosity and the most generous states are not in every case the most likely suspects. States that are generous in some programs are less generous in others and it appears that there may be a trade-off between making it easier for individuals to access benefits (eligibility rules) and the amount they receive (replacement rates).