Year of Publication



Martin School of Public Policy and Administration

Executive Summary

Since the Financial Crisis of 2008, pension plans in the United States have sunk into a crisis that there seems to be no escape from. While all 50 states have at least one retirement system for employees, several of these systems have been repeatedly hit by massive problems involving funding for future benefits and paying out current obligations. Taken as a collective whole, public retirement systems in this country have been estimated to be underfunded anywhere from $934 billion to $3.2 trillion. Several states barely have half the money needed to meet all obligations promised. While several studies have been conducted to establish the reasons for these numerous problems affecting retirement systems in the United States, more attention should be paid regarding the impact of politics and partisanship on state retirement policy as a whole, and state retirement system funding, more specifically. This study offers such an analysis.

Using the funded ratio as the primary measure of retirement system health, and the most recent pension plan data available, a panel dataset spanning the years 2001- 2015 was created to investigate the relationship between partisanship and pension plan funding for state-administered retirement systems. This panel data was then tested via statistical analysis to determine these potential relationships. An intensive statistical model containing four fixed-effects regressions test the funded ratio with 16 explanatory variables. This set of explanatory variables includes several variables representing partisan political control to properly conduct the analysis.

The results derived from the regressions present highly unexpected and interesting results. Contrary to the expected outcome, divided government has a positive, and significant, impact on the funded ratio of state-run retirement systems. Divided state government is a 0.018 increase to the funded ratio. Other unexpected results include no effect on the funded ratio from a Democratic governor or Democratic controlled state senate, and a negative effect on the ratio from a state house controlled by Democrats and the particular number of seats controlled by Democrats in a state house. The percentage of seats held by Democrats in a state senate has a positive and statistically significant effect on the funded ratio via a 0.28 increase.

Following from the results of the fixed-effects regression analysis, the author recommends both political parties undertake evidence-based approaches to retirement system policy to ensure that funding for pension obligations is at acceptable levels. Attention should always be paid to retirement system policy no matter what party is in control, but it is of vital importance when the state is operating under divided government. It is also recommended that states should study and learn from the states that have achieved better pension policy outcomes under divided government to learn best practices and solutions to combat the underfunding problems and keep up with all current and future obligations.