Estimating state expenditures on public education : an econometric study from pooled time-series and cross-sectional data

This study focuses on the estimation of state spending on public education. The purpose of this study is to review previous empirical studies, to formulate a demand model for state public education based on the median-voter approach, and to estimate the demand equations for public education by region and for the U.S as a whole, using pooled time-series and cross-sectional data from 1982 to 1991. The empirical results of this study reveal several findings: first, the hypothesis testing has verified that the estimation of the demand for public education by region is justifiable. Second, empirical evidence indicates that the dynamic model based on a partial adjustment mechanism pertaining to spending in public education is more desirable than the static one. Third, the GLS method based on a cross-sectionally correlated and time-wise autoregressive model provides better results for analyzing the demand for public education from state government than do the MLE or OLS methods. Fourth, the independent variables for empirical estimation are generally correct with their signs and are consistent with the theoretical expectations. Fifth, the impacts of public educational policies on state spending for public education are examined. Finally, the possibility of achieving complete wealth neutrality by adjusting federal aid is discussed based on the estimated tax-price elasticities.