Association for Public Policy Analysis & Management

Major economic, political, and social events open the window for dramatic shifts in policy.  They draw attention to new conditions, shift awareness to previously ignored problems, and provide the opportunity to reframe existing policies and issues.  In doing this, they sometimes lead to policy punctuations—major expansions or contractions of programs, substantial reworking of policies, or the initiation of new policies or programs (Jones and Baumgartner, 2005). The Great Recession of 2007-2009, whose effects linger even today, was a substantial dislocation that set the stage for significant policy initiatives at the federal, state, and local levels in the United States. It created the opportunity for – but by no means the certainty of – substantial changes in social welfare programs. 

In this paper, we seek to answer questions about changes in the rules governing state TANF programs during the Great Recession.  Did American state programs become more or less stringent in response to the recession? What accounts for changes in welfare rules among the states? In the context of the move toward more stringent welfare policies in the American states that began in the 1980s and growing economic inequality, what affected state responses?  In particular, did variations in levels of economic inequality across the states shape changes in the rules.  Did states with higher levels of inequality move in a more restrictive direction or less restrictive direction than states with lower levels of inequality?  The analysis focuses on 43 rules identified by De Jong et al. (2006) as components of three dimensions of welfare policy in the immediate aftermath of the adoption of the Personal Responsibility and Work Opportunity Reform Act of 1996 and four rules Soss, et al. (2001) identified as critical for access to welfare benefits. To account for changes in the rules, we incorporate a variety of social, economic, and political variables, including the degree of income inequality, in the analysis.  We estimate event history (hazard) models for rules changes in the fifty states between three points in time:  2008, 2010, and 2012.