US states have many policy options to promote short-term and long-term development of local economies and workforces. During the 1990s and 2000s, Alabama voters rejected a number of policy proposals designed with long -term, balanced development in mind, maintaining by default their state's capital-subsidy framework for economic growth. These outcomes were surprising; many observers expected the bills to pass because they were progressive and had strong business support. Using a power framework and methodology developed by Gaventa (1980) and others, this mixedmethods study explores the way power was used to shape two vote outcomes. Through consideration of the historical, structural, and cultural makeup of the state, and mindful of theories about -the end of Southern exceptionalism[DOUBLE VERTICAL LINE] (Shafer & Johnston, 2006), the study develops hypotheses about the continued role of regressive institutions in shaping Alabamians' ideas about development. Content analysis of public debate suggests that -jobs and development[DOUBLE VERTICAL LINE] arguments for the proposals were less resonant with voters than expected. Subsequent regression analysis of countylevel voting confirms that race still plays an important role in predicting voters' response to progressive development policies. This research is important because it focuses on the causes and durability of spatial inequality at the subnational level, a scale that has been neglected by social science but is a key site for public policy interventions (Lobao, Hooks, & Tickameyer, 2008). It also provides a lens for the iii study of economically -uneven[DOUBLE VERTICAL LINE] regions negotiating post-industrial change and a new knowledge-based global economy. iv