Last year, a report by the Pew-MacArthur Results First Initiative, States’ Use of Cost-Benefit Analysis, evaluated the use of Cost-Benefit Analysis by states. Among the questions it asked were, "Are states conducting cost-benefit analyses?" and "Do they use the results when making policy and budget decisions?"
On pages 13-14 of the full report, an interesting map illustrates that "Ten states—Florida, Kansas, Minnesota, Missouri, New York, North Carolina, Utah, Virginia, Washington, and Wisconsin— were among the leaders in at least two of the rating criteria, making them national leaders on the use of cost-benefit analysis in policymaking."
On the other hand, "...11 states lagged behind their peers, producing few studies, making little effort to evaluate multiple alternatives, and reporting very little meaningful policy or budget impact from their analyses."
Along with my student Rahul Patel, I analyzed data from this Pew Report. We merged demographic and socioeconomic data on states from the City and County Databook, and presidential voting data from USA Counties, and estimated simple and multiple regression models. The following three tables summarize our data and results:
Population and education are statistically significant determinants of BCA use across all models. Surprisingly, larger states are less likely to use BCA. States with more college-educated residents use BCA more, as expected, but the income variable is generally insignificant and the estimated sign is not robust across models. Finally, whileliberal political ideology is positively correlated with BCA use, (and this may or may not be as expected, depending on the perspective of the reader) this finding is not robust to socioeconomic and demographic controls.
These socioeconomic and demographic controls by themselves explain about a third of the variation in the dependent variable, and the inclusion of political measures does not improve the fit of the model.
On pages 13-14 of the full report, an interesting map illustrates that "Ten states—Florida, Kansas, Minnesota, Missouri, New York, North Carolina, Utah, Virginia, Washington, and Wisconsin— were among the leaders in at least two of the rating criteria, making them national leaders on the use of cost-benefit analysis in policymaking."
On the other hand, "...11 states lagged behind their peers, producing few studies, making little effort to evaluate multiple alternatives, and reporting very little meaningful policy or budget impact from their analyses."
Along with my student Rahul Patel, I analyzed data from this Pew Report. We merged demographic and socioeconomic data on states from the City and County Databook, and presidential voting data from USA Counties, and estimated simple and multiple regression models. The following three tables summarize our data and results:
Population and education are statistically significant determinants of BCA use across all models. Surprisingly, larger states are less likely to use BCA. States with more college-educated residents use BCA more, as expected, but the income variable is generally insignificant and the estimated sign is not robust across models. Finally, whileliberal political ideology is positively correlated with BCA use, (and this may or may not be as expected, depending on the perspective of the reader) this finding is not robust to socioeconomic and demographic controls.
These socioeconomic and demographic controls by themselves explain about a third of the variation in the dependent variable, and the inclusion of political measures does not improve the fit of the model.