Pensions in the Trenches
How Pension Spending Is Affecting U.S. Local Government
Sarah F. Anzia (University of California, Berkeley)
Local government budgets are in the spotlight. The COVID-19 economic downturn has decimated certain streams of local government revenue. Scrutiny of policing has raised attention to municipal expenditures. It might seem like public employee pensions are disconnected from all of this: they are usually discussed as a state-level issue, and one involving quantities like funding ratios, unfunded liabilities, and investment returns rather than spending. But the reality is that they are very much connected. Spending on public employees’ retirement benefits—including those of public safety employees—is an important part of local government budgets everywhere. And long before 2020, many experts were warning that pension costs were on the rise, forcing changes to how local governments operate.
Until now, it hasn’t been possible to say how pronounced or widespread any of these changes have been. My article in Urban Affairs Review brings together data on the pension expenditures of over 400 municipalities and counties from 2005 to 2016 to examine how local pension expenditures have changed and how that is affecting local governments.
First, the basics of the issue at hand: About 14 million people work full time for U.S. state and local governments, and nearly all of them are eligible for a defined-benefit pension. Most are enrolled in state-operated pension plans like CalPERS, but there are also many local governments that operate their own plans. Pension benefits are supposed to be prefunded, with government employers and employees setting aside funds to pay for the retirement benefits earned each year. But over the last few decades, officials have both made pension benefits more generous and underfunded the benefits that have been promised. As a result of these trends and the investment losses of the Great Recession, U.S. public employee pensions as of 2017-2018 were underfunded by somewhere between $1.24 trillion and $4 trillion, depending on the estimates used.
State and local governments have therefore been called on to make larger pension contributions—to pay for the more generous benefits and make up for funding shortfalls. Yet because of a lack of appropriate data, researchers haven’t been able to evaluate how this is affecting local governments. There are rich datasets of state and large local pension plans, but data on pension plansdon’t tell us about the pension expenditures of particular local governments, most of which contribute to multiple pension plans.
I assembled a new dataset to evaluate how local governments’ pension spending has changed over time and how the governments have adapted. Over the course of about two years, with the help of a team of excellent research assistants, I collected 12 years of comprehensive annual financial reports (CAFRs) from hundreds of local governments across the United States. From each CAFR, we pulled out information on what the local government contributed to each of its employee retirement plans in that year.
The resulting dataset shows that between 2005 and 2016, pension expenditures grew in real terms in 88% of the municipalities and counties. The median within-government growth over this 12-year period was 56%, but there is a long right tail to the distribution. For instance, in 26% of the cities and counties, inflation-adjusted pension expenditures more than doubled in 12 years. I also find that in most cities and counties, pension expenditures increased as a share of general revenue, as did pension expenditures per local government employee. For example, the median within-government change in pension expenditures per employee from 2005 to 2016 was $1,419, and in 25% of the cities and counties, per-employee pension expenditures increased by $3,542 or more.
These, by themselves, are important findings. For the first time, these data allow us to see how pension expenditures have changed in a large number of local governments of different sizes across the United States. But an important next step is to examine whether there are any trends in how local governments have responded to those changes. My approach is to evaluate whether there are discernable trends in cities’ and counties’ responses—and any clear links between those trends and changes in local pension spending. In particular, I focus on whether larger pension spending increases are associated with larger increases in revenue, decreases in employment, or decreases in spending on items other than retirement benefits.
I do not find any link between rising pension spending and increasing general revenue. In cities and counties that experience greater-than-average increases in pension spending per employee, the next year does not bring greater-than-average increases in revenue.
Instead, rising pension expenditures are associated with reductions in public-sector employment. On average, a 10% increase in a local government’s pension expenditures per employee is associated with a 0.44% decrease in its employment the following year. I also find that this relationship between local government pension spending and employment reductions is more pronounced 1) in cities and counties with collective bargaining for public-sector workers, 2) for non-public safety employees than for public safety employees, and 3) in states with restrictive tax and expenditure limitations.
The results also provide new evidence that the politics of pensions cuts across standard ideological and partisan lines. One might think more Democratic, liberal cities and counties would be more likely to respond to pension expenditure increases by increasing revenue and staving off employment reductions. But they are not.
The coming months and years will involve important conversations about how local revenue and spending will look going forward, and clear-eyed discussion of public pensions needs to be a part of them. Well before the COVID-19 crisis, city and county pension expenditures were rising. And if there is a trend in how local governments have responded, it’s not that they’re increasing revenue—it’s that they’re cutting back their workforces. The implications of this are significant not only for local employees but also for everyone who depends on local government service provision.
Sarah F. Anzia is the Michelle J. Schwartz Associate Professor of Public Policy and Political Science at the University of California, Berkeley. She is the author of Timing and Turnout: How Off-Cycle Elections Favor Organized Groups (The University of Chicago Press, 2014) and has written widely about state and local government and the politics of public pensions.