How do State Revenue Swings Influence the Provision of Health, Education, and Welfare Services?

The dramatic deterioration in state finances during the Great Recession of 2008-2009 raised concerns regarding the ability of government to support community health and education. We analyze changes in state spending for all contiguous states from 1989 through 2009 to examine how states balance budgets in response to revenue shortfalls. We measure revenue shortfalls using revenue swings -- changes in revenues per capita from existing state sources (inflation adjusted). We derive a reduced form public finance model that enables us to characterize the effect of state revenue swings on health, welfare, and education, while controlling for factors that affect the demand for these services. We provide evidence that higher deficits lead to temporary cuts such as reductions in Medicaid benefits for children, with short term and long term negative consequences.

            We use several dependent variables in our analysis, including state spending on Medicaid, education, and other welfare. We further characterize states’ Medicaid spending, examining the determinants of enrollment, spending per capita, and the percent of teen mothers with low-weight births.  We also examine monthly AFDC/TANF benefits and Head Start enrollment. We analyze K–12 state spending, the pupil to teacher ratio, and a range of state higher education spending and outcomes measures. We draw data from multiple sources, including the U.S. Census, the National Association of State Budget Officers, the U.S. Department of Health and Human Services, the National Governors Association, and the Bureau of Economic Analysis.

            Our main explanatory variables are yearly revenue swings and lagged balances. Negative revenue swings capture the immediate effect of a revenue shortfall on state spending, while a weighted sum of lagged balances (based on a formula derived in our paper) captures the long-term effect of past revenue swings on spending. We control for the potential endogeneity of revenue swings and expenditures in two ways. First, we estimate the model in first differences to remove state fixed effects. Second, we remove legislated changes in revenues from our measure of revenue swings. We show that variability in revenue swings is driven by the responsiveness of  varying taxation methods across states (e.g. progressivity of income tax rates or level of sales taxes) to upturns or downturns in the economy.

            We find that cuts in state revenues reduced spending on Medicaid, education, and public safety, comprising 19%, 33%, and 5% of state general expenditures in 2008. These cuts resulted in fewer children enrolled in Medicaid, lower spending per child eligible for Medicaid, more low-weight births, higher pupil to teacher ratios, higher college tuition, and fewer associate degrees awarded. State spending cuts did not lower benefits for AFDC/TANF, or lower Head Start enrollments. Many individuals may benefit from gradual increases in public spending that occur as a result of rising income per capita in the long run. However, the increase in low birth weights and reduction in education outcomes that occurred during the Great Recession may have long-term negative consequences for the cohort of individuals needing immediate assistance during that period.