Special CCICADA/DIMACS Seminar

Title: Policy Tool Bundling: Predicting the Selection of Policy Instruments Using a Bayesian Multivariate Probit Model

Speaker: Anthony J. Kassekert, Dept. of Homeland Security and Florida State University

Date: Thursday, March 28, 2013 1:15 - 2:15 PM

Location: CoRE Bldg, Room 431, Rutgers University, Busch Campus, Piscataway, NJ


Local governments have a vast array of economic development policy options available to spur growth and/or provide political patronage to local constituencies. In theorizing why governments choose a particular incentive over another, the policy sciences have developed general theories for these incentives including the policy tools framework and policy diffusion. Both theories have generated insight into the individual determinants of the policy selection but have overlooked interdependencies between policy options and why they may be bundled together. This research builds on the political science theories of diffusion and policy tools to explain why multiple tools are used in conjunction with one another to solve public problems. I theorize that policy bundling is strategic by local officials, incentivized by political institutions, and heavily influenced by previous policy decisions.

The theory of policy tool bundling is empirically tested using panel data from the state of Georgia. The presence of policy tool bundling is assessed by modeling four economic incentives simultaneously with a multivariate probit (MVP) model estimated using Bayesian methods. A MVP model allows for a series of binary outcomes models to be correlated with each other. The correlations are used as indications of policy bundling. The results demonstrate that bundling is occurring between free or reduced cost land and expedited permitting and also between free or reduced cost land and industrial development bonds. No evidence of bundling was found between other incentives indicating that while policy bundling does occur in economic development, many of the observed relationships between policies are not strategic.

No knowledge of policy theories or statistical methods will be expected prior to attending. The presentation will be applied and will exhibit all requisite information. Questions and feedback are welcomed.


Anthony Kassekert is a Courtesy Professor with Florida State's Institute for Energy Systems, Economics, and Sustainability (IESES) and a statistician with the Department of Homeland Security's Office of Strategy, Planning, Analysis, & Risk. His professional interests include immigration policy, performance measurement, sustainable energy policy, federalism studies, and applied statistics.