Marlowe, JustinLeiser, Stephanie Ann2014-10-132014-10-132014-10-132014-10-132014Leiser_washington_0250E_12938.pdfhttp://hdl.handle.net/1773/26029Thesis (Ph.D.)--University of Washington, 2014States' use of tax incentives to lure business investment has drawn a lot of attention from policymakers, the public, and academics. Despite mixed assessments of their efficacy in creating net increases in jobs and investment, business tax incentives continue to be highly popular among state policymakers. This dissertation draws on theories of policy diffusion and tax competition to explore why states adopt tax incentives for business, with a focus on how states influence each other in their policy choices. The first essay is a mixed methods case study of the spread of incentives for film and video production. It examines qualitative data on interviews with state policymakers, as well as quantitative data on the adoption of film incentives in all 50 states. The analysis suggests that state adoptions of film incentives are primarily driven by two factors: the size of the existing film industry in the state and a "bandwagon" effect based on the total number of adopters. The second essay investigates whether similar adoption patterns hold for four other state tax incentive policies: Investment Tax Credits, apportionment formula changes, R&D tax credits, and Job Creation Tax Credits. The quantitative event history analyses show that factors that influence adoption decisions are largely inconsistent across the different incentive types, but the evidence is consistent with the idea that the adoption of business tax incentives is a zero-sum game or "race to the bottom." One theme that emerges in both of the first two essays is the importance of modeling and interpreting how diffusion processes change over time. The third essay is a methodological discussion of duration dependence--how the hazard of adoption changes over time--in the context of policy diffusion. It argues that the most commonly used methods for modeling diffusion are inadequate for detecting certain system-level diffusion dynamics and should be complemented by a more thorough analysis of duration dependence. It discusses how duration dependence relates to other methodological issues and provides a list of recommendations for researchers on how to properly model and interpret duration dependence in quantitative policy diffusion studies.application/pdfen-USCopyright is held by the individual authors.event history analysis; policy diffusion; state policy; tax policyPublic policyPolitical Sciencepublic affairsThesis