Files
Abstract
This dissertation assesses how changes in discretion level affect welfare programs, by developing program-level measures of discretion and examining its moderation effect on welfare spending. States’ discretion regarding social safety net programs has been changed by welfare reforms in the U.S., and these federal acts provide rich information on discretion, describing not only legislative goals but also delegated agents’ rights, responsibilities, expectations, and prohibitions when implementing policies or programs. This research investigates eight major federal acts on welfare reform since 1935 affecting Aid to Families with Dependent Children (AFDC), and its successor, Temporary Assistance for Needy Families (TANF). Findings indicate that TANF allows more discretion for state governments under less federal oversight, and that co-operations of diverse policy tools have produced wide state-level variations.The new measures of states’ discretion in program administration come through analyzing the provisions of 18 federal welfare reform acts using the Natural Language Processing (NLP) method, specifically the spaCy library in the Python language. Findings show that the level of discretion, rather than steadily changing in one direction as assumed, has increased by different degrees or remained static in response to these federal acts.
Finally, this research empirically estimates the moderating effect of discretion on TANF expenditures, by fiscal condition, political preferences, and racial diversity of state governments. Results reveal that discretion itself does not directly change the spending, but the effects of key variables of interest significantly change when discretion increases: discretion is a moderator in the relationship between characteristics of state governments and welfare expenditure. Taken together, this dissertation develops analytic methods to assess discretion from legal statements, allowing for its measure in various policies and programs. Furthermore, the moderating role of discretion in decision making on governmental assistance offers a different perspective with which to resolve discrepancies in empirical research on welfare state theory.