Cross-state disparities in pharmaceutical expenditure in the U.S.

This paper studies the existence of pharmaceutical expenditure disparities among US states and the District of Columbia for a sample that covers 1991–2020 using State of Residence data. The results obtained reject the existence of a single pattern of behaviour in regional pharmaceutical expenditure across the US. Instead, we identify the existence of four convergence clubs. Mixed results were obtained when we disaggregated the data into public and non-public expenditure. We cannot reject the null hypothesis of convergence for public expenditure, but we can reject it for non-public expenditure. The use of a set of local and global statistics to test for spatial autocorrelation show that states with similar expenditures are spatially clustered for total and non-public pharmaceutical expenditures. An analysis of the factors driving the formation of convergence clubs shows a varied picture. For the total pharmaceutical expenditure per capita, we obtained that the unemployment rate, the percentage of females, and physical activity play an important role in explaining the convergence club creation. However, chronic conditions and race also play a crucial role in non-public expenditure.