Government Expenditure on the Public Education System

Working Paper: NBER ID: w26425

Authors: Chao Fu; Shoya Ishimaru; John Kennan

Abstract: We investigate equilibrium impacts of federal policies such as free-college proposals, taking into account that human capital production is cumulative and that state governments have resource constraints. In the model, a state government cares about household welfare and aggregate educational attainment. Realizing that household choices vary with its decisions, the government chooses income tax rates, per-student expenditure levels on public K-12 and college education, college tuition and the provision of other public goods, subject to its budget constraint. We estimate the model using data from the U.S. Using counterfactual simulations, we find that free-public-college policies, mandatory or subsidized, would decrease state expenditure on and hence the quality of public education. More students would obtain college degrees due to increased enrollment. Over 86% of all households would lose while about 60% of the lowest income quintile would gain from such policies.

Keywords: Public Education; Free College; Human Capital; State Expenditure

JEL Codes: I00; J01; J24


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
Mandatory free public college policies (H49)Decrease in state expenditure on K-12 and college education (H75)
Decrease in state expenditure on K-12 and college education (H75)Decline in educational quality (I21)
Mandatory free public college policies (H49)Decline in educational quality (I21)
State governments raising tax rates (H71)Reduction in per-student expenditures (H52)
Overall college enrollment increases (I23)Graduation rate in public four-year colleges declines from 62% to 58% (I21)
Mandatory free public college policies (H49)Loss in welfare for approximately 86% of all households (H53)
Mandatory free public college policies (H49)Gains in welfare for about 60% of the lowest income quintile (D69)

Back to index