In 2011, the U.S. government spent $940 billion on means-tested welfare assistance – much more than what it spent on public education and national defense. Of that $940 billion, half went to the TANF program. And yet, welfare caseloads have declined steadily since the passage and implementation of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996, with national TANF family caseloads decreasing by 2,512,762 (58 percent), and individual caseloads falling by 7,792,314 (64 percent), between 1996 and 2005. Numerous studies have credited economic conditions with most of the decline in TANF caseloads, but this study finds an important weakness in research primarily focused on economic factors: their inability to explain specifically, in terms of the relevant operating systems and mechanisms, how economic factors led to a decline in TANF caseloads. As poverty and unemployment rates decreased from 1996 to 2000, the need for TANF assistance also declined, leading to a consequent decline in the number of families on TANF. That said, the data from 2003 reveals the insufficiency of previously proposed economic explanations, and so, despite enjoying some support in the literature, it is premature to conclude that economic conditions were the main cause of the decline in TANF caseloads. Further verification is required.

This study argues that attributing a decline in TANF caseloads to a strong economy is a politically useful claim for the U.S. government. If TANF caseloads decrease –whether the economy is growing or in recession – the decline can be regarded as a major success. If TANF caseloads increase as the economy improves, the government can declare that the provision of generous governmental assistance to TANF recipients certainly will lead to an increase in TANF caseloads. On the other hand, when the economy is in recession, the government can claim that the increase in TANF caseloads should be attributed to the weak economy. As a result, while the TANF subsidy per capita ratio and administrative expenditure ratio remain constant, if decision makers wish to suppress the number of TANF cases, one of the most feasible methods is to enforce criteria rigidly and increase the ratio of TANF applications that are rejected, and consequently remove applicants and recipients from the rolls.

In order to challenge the strong economic factor argument, this study intends to establish an innovative model to better explain TANF caseload changes. Using variables based on official aggregate data collected from 2000 to 2008, this study first utilizes Logit Regression analysis to draw out secondary variables that best explain changing welfare caseloads among the various states; then, these variables are run through LISREL analysis to build a model that best explains the factors influencing changes in welfare caseloads among the various states. Unlike previous studies, this analysis finds that no matter which factor is highlighted, it is the administrative behavior factor, followed by the economic, fiscal, social needs and political factors, that directly influence changes in TANF caseloads. This study demonstrates that TANF caseload decline is directly affected by neither fiscal expenditures nor economic factors, and concludes that the TANF administrative effect factor exerts a pivotal and direct influence on the ratio of TANF families, both theoretically and practically.