Urine or You’re Out: Racialized Economic Threat and the Determinants of Welfare Drug Testing Policy in the United States, 2009–2015

ABSTRACT

Since 2009, fifteen U.S. states have passed some sort of restrictive drug-testing program for welfare-applicants and/or suspected recipients. Sociologists are often concerned with identifying the conditions under which states implement a particular social policy. As such, we ask what conditions separate states that have implemented drug testing policies for welfare applicants from those that have not. Using discrete-time event history analysis models we test several existing theories of policy creation, namely group threat and partisan politics. Results indicate that declines in white labor force participation, and possessing a Republican governor, are positively associated with the passage of welfare drug testing policy. Contrary to explanations given by political operatives, neither state fiscal conditions, nor welfare caseloads, are statistically associated with passage. Our results re-affirm the association between racial and economic threat and reactionary social policy, as well as underscore the importance of political power in implementing such policy. This deepens and broadens our understanding of the longstanding American animosity towards welfare and welfare recipients, and, most importantly, the ongoing shift towards disciplinary poverty management in the United States, even in the face of a substantial economic downturn.

KEYWORDS:

• Social policy
• welfare policy
• political sociology
• poverty management
• racial threat
• event history analysis

Acknowledgments

We would like to thank Keith Bentele, Ron Breiger, Terrence Hill, Anna Jacobs, Erin Leahey, Simone Rambotti, Kathleen Schwartzman, Julia Smith, the participants of the Arizona Inequality Workshop, and the Social Welfare Programs session at the 2016 American Sociological Association Meeting for their comments and suggestions.

Notes

1. Caseloads can also be lowered by related state policies, such as whether a state has a supplemental Earned Income Tax Credit (Loprest 2012).

2. Alaska and Hawaii have been excluded from our analysis. This is a common procedure in U.S. state-level social policy analysis because these two states calculate poverty in unique ways.

3. See http://www.ncsl.org/research/human-services/drug-testing-and-public-assistance.aspx#2015 Legislation.

4. NCSL claims that Arizona passed their requirement in 2011, although several reputable independent sources indicate that the state passed their requirement in 2009 (Kreig 2015, U.S. Department of Health and Human Services).

5. An additional reason for excluding unemployment rate (as well as poverty rate) from our analysis is that they have a high degrees of collinearity with “percent of whites out of labor force” (0.62 and 0.68).

6. In additional analysis we include a measure of the percentage of a state’s population receiving SNAP (food stamps). This measure is highly correlated with one of our key theoretical predictor variables (0.67) and is therefore excluded from our modeling due to collinearity.

7. To be sure, it is often not in calendar-year intervals, but the overall point remains.

8. We are not implying that the Great Recession was the beginning of declining white labor force participation. This had been occurring for years; however, the Great Recession triggered a substantial exit, particularly among younger and older white men.

9. We have adjusted the date of Arizona’s bill to reflect our independent research.