Search

This article employs 1963-97 panel data for the 48 contiguous U.S. states (and District of Columbia) to examine the relationship between real personal income and real education expenditures as well as that between real personal income and six measures of real research and development expenditures. Bivariate regressions are employed to determine whether the information content between real education expenditures and real income runs from real income to real education spending or vice versa. The authors find that when data are relative to the U.S. average, the direction of information content runs from real state-level education expenditures to real state-level income. (JEL 12, H72), I. INTRODUCTION State and local policy makers face a myriad of competing demands for the scarce resources available for allocation. Among programs competing for funding that the authors are interested [...]

The utility of agents in both stationary non-rational expectations equilibrium and stationary rational expectations equilibrium were examined in an overlapping generations model. Rational expectations models assume that utility will be enhanced if information is used efficiently and expectations are not biased. It was suggested that when government expenditures are sourced from a distortionary tax, non-rational expectations equilibrium is Pareto superior to rational expectations but not Pareto optimal.