State‐level austerity, education, and large urban labor markets: Evidence from fiscal policy experiments in Kansas and Wisconsin

Abstract

In 2011, Kansas and Wisconsin experimented with fiscal austerity policies. Proponents of the tax and public expenditure cuts argued that they would spur economic growth, while opponents claimed that it would harm vulnerable populations. I use the timing of these policies and the synthetic control method to examine how state‐level austerity affected earnings per worker for high and low education segments of the workforce in major urban areas. The results suggest that the Kansas and Wisconsin fiscal experiments hurt low education workers in some urban areas. The estimates also show heterogeneity in the treatment effect. Notably, Madison, WI, was the only urban area that experienced substantial positive growth in both education segments.

Estimating the economic effects of US state and local fiscal policy: A synthetic control method matching‐regression approach

“…Own-source revenues have adverse effects on most outcome variables. Consistent with SCM case studies of single policies , 2020, reducing individual income taxes does not appear to stimulate economic growth and more likely reduces growth relative to other state and local taxes or expenditures. The property tax is the most likely tax to negatively affect state economic performance.…”

“…The type of state and local tax appears to matter more than the category of expenditures, in which the property tax is the most likely tax to negatively affect state economic performance. Consistent with SCM case studies there is an absence of evidence that reducing individual income taxes benefits state economies relative to other taxes and expenditures , 2020. Highway expenditures have a positive effect on both employment and per capita GDP that dominates the negative own-source revenue or property tax effects.…”

“…Rather than focus on a single economic indicator, we follow Wang and Rickman (2018) and use principal component analysis of several outcome indicators to derive a narrative of state growth experiences such as whether it derives more from demand versus supply. We use the principal components to match with SCM instead of matching each economic outcome variable separately , 2020. The patterns of correlations of the outcome variables with the principal components are the basis for the interpretation of the channel of economic influence of the fiscal variables.…”

“…We follow the integrative analysis of Wang and Rickman (2018) and use principal components to facilitate the interpretation of the patterns among multiple measures of economic outcomes. This avoids having each outcome variable producing a different counterfactual comparison (e.g., , 2020. We also use principal components to reduce the dimensionality of the control variables rather than use model selection criteria (Reed, 2008).…”

“…The use of all states in a pooled sample though involves comparisons of those with highly differing characteristics which can create interpolation bias (Abadie et al., 2010, 2015). To overcome the limitations of conventional regression analysis, several state and local fiscal studies have used the synthetic control method (SCM) matching approach (Komarek, 2020; Lusch & Stekelberg, 2020; Mughan & Propheter, 2017; Rickman & Wang, 2018, 2020; Spreen, 2018). SCM attempts to avoid the interpolation bias of conventional regression analysis by constructing counterfactual comparisons based on affinities between the states (Abadie et al., 2010, 2015).…”