Breaking Bad: Public Pensions and the Loss of That Old-Time Fiscal Religion.

• Author: Smith, Daniel J.

Despite maintaining strict balanced-budget requirements, U.S. state and local governments have accumulated trillions in unfunded pension liabilities (National Conference of State Legislatures 2010). (1) These unfunded liabilities represent a growing economic and budgetary concern for many state and local governments (Brown and Wilcox 2009; Ricketts and Walker 2012; U.S. State Budgets 2012; Kiewiet and McCubbins 2013). Although the extent of these unfunded liabilities varies according to assumptions, estimates generally range between $3 and $5.2 trillion (Biggs 2016; Foltin et al. 2017, 2018; Thornburg, Komissarov, and Rosacker 2017; Norcross and Gonzalez 2018). Using the latest available data from 2017 and a discount rate of 3 percent, the Stanford Institute for Economic Policy Research estimates these state and local pension liabilities to be $5.176 trillion. (2)

Many distal factors--such as liberal discount rates (Gold and Latter 2009; Waring 2012; Naughton, Petacchi, and Weber 2015), unsound investment policy (Ryan and Fabozzi 2002; Lucas and Zeldes 2009; Stalebrink 2014), accounting and disclosure policies (Marks, Raman, and Wilson 1988; Vermeer, Styles, and Patton 2012; Davidyan and Waymire 2018), and inadequate funding leading up the financial crisis (Munnell, Aubry, and Quinby 2011)--have been identified as causal explanations for this growth in unfunded pension liabilities.

More proximate factors, however, such as public-choice explanations, have been contested in the literature (Marks, Raman, and Wilson 1988). Public-choice theory makes symmetric behavioral assumptions across government and economic actors, assuming that both pursue their self-interest, broadly considered, and respond to the institutional incentives they face. In other words, once elected or appointed, policy makers do not suddenly switch from operating in a self-interested fashion in the market to operating in the public interest in government. More broadly, public choice examines government procedures and outcomes through the lens of economics. Whereas Olivia Mitchell and Robert Smith (1994), Cynthia Sneed and John Sneed (1997), Dashle Kelley (2014), and Steven Thornburg, Sergey Komissarov, and Kirsten Rosacker (2017) find evidence supporting public-choice explanations, including the median voter and special-interest-group models, Marguerite Schneider and Fariborz Damanpour (2002) find limited support for public-choice explanations.

This paper contributes to this literature by providing a previously overlooked public-choice explanation for the growth of unfunded public-pension liabilities: the undermining of the old-time fiscal religion. According to this theory, taxpayer willingness to pay or not pay taxes plays an important, albeit imperfect, role in signaling taxpayer assessment of government expenditures. This taxpayer calculus provides a check on government spending because taxpayers, bearing the full burden of current government expenditures, have more incentive to closely monitor and restrain the growth of government expenditures. The signal is most effective when governments maintain balanced budgets--that is, follow the "old-time fiscal religion"--because this ensures that spending and taxes are contained to the current generation.

The intergenerational transfer of spending and taxes undermines taxpayer constraint on the growth of government because future taxpayers are incapable of providing an assessment of that growth and because current taxpayers have a reduced incentive to monitor spending promises (and the taxes required to support them) because they will fall on future generations (Kotlikoff and Burns 2012). Importantly, once the old-time fiscal religion is discarded and taxpayers have less incentive to monitor expenses, government actors are better able to minimize resistance to the promises of increased retirement to public employees made through the use of inappropriate actuarial techniques that generate the fiscal illusion that these promises are funded (Buchanan 1999a, chap. 10). Fiscal illusion, by making it difficult for taxpayers to accurately assess the projected costs of the increase in retirement benefits, can enable policy makers to deliver benefits to public employees, a concentrated and often well-organized special-interest group, at the expense of future taxpayers.

Public-choice theorists first used the old-time fiscal-religion theory to explain the substantial growth in deficit spending at the federal level following the Keynesian revolution. By removing the balanced-budget tradition, Keynesian economics helped undermine taxpayer constraint on the growth of government (Buchanan and Wagner 2000). Removing this constraint thereby permitted policy makers to utilize fiscal illusion more effectively, often in the form of hidden taxes or public debt, to maximize their spending (Buchanan 1999a, chap. 10).

This paper argues that the broad use of the defined-benefit model in public-pension plans has, like the rise of deficit spending under Keynesianism, undermined the old-time fiscal religion at the state and local level and, in combination with fiscal illusion and the argument that public pensions contribute to stimulating state and local economic growth, has led to substantial growth in unfunded pension liabilities. By enabling policy makers to push the costs of promises made to current employees onto future generations, defined-benefit public pensions have undermined taxpayer constraint. Thus, this paper complements the existing literature by providing an additional and overlooked public-choice explanation for the accumulation of unfunded pension liabilities.

Controlling the growth of unfunded pension liabilities would require restoring taxpayer constraint (Giertz and Papke 2007). One potential way to restore taxpayer constraint would be to transition new public employees to defined-contribution retirement plans. Defined-contribution retirement plans, by requiring upfront funding, would help prevent the intergenerational transfer of spending and taxation. Many states have taken steps toward reforming their pension systems either by lessening benefit generosity, tightening up their retirement provisions, or replacing their defined-benefit plans with either a hybrid defined-benefit/defined-contribution plan (3) or a pure defined-contribution plan as a solution to the large funding shortfall in pensions (Snell 2012; Ali and Frank 2018; Gorina and Hoang forthcoming).

The first section of the paper provides a brief discussion of the history of pensions in the United States. In the second section, we briefly review the public-choice arguments for how Keynesian economics undermined the old-time fiscal religion. In the third section, we argue that defined-benefit public pensions have undermined the old-time fiscal religion in U.S. state and local governments. The paper closes by explaining how state and local governments can help revive the old-time fiscal religion by transitioning to defined-contribution retirement plans.

A Brief History of Pensions in the United States

The history of public pensions dates back to antiquity when rulers motivated their militaries with the promise of lifetime support, especially for those disabled in combat.

Since then, many nation-states have adopted this practice of establishing pensions for military personnel. The Continental Congress created the first pension plan for navy personnel in 1775, financed by the sale of prizes seized by the Continental navy (Cogan 2017, chap. 3). Military pensions quickly became an expected entitlement for U.S. veterans, with the tendency for these programs to be expanded incrementally on a regular basis, especially during good economic times or for electoral gain, but with drastically underestimated costs (Costa 1998, chap. 8; Cogan 2017, 23, 40). For instance, with the Revolutionary War pensions, Congress "established unfortunate precedents by bailing out an insolvent navy fund with an annual infusion of general revenues and by using creative accounting procedures to mask the use of general revenues to finance pensions rather than earnings on trust fund assets" (Cogan 2017, 25).

These military-pension systems were extended to public employees during the late nineteenth and early twentieth centuries. Beginning in 1857, cities in the United States began to offer pensions for their nonmilitary employees, and in 1911 states began to follow this path (Clark, Craig, and Wilson 2003, 167). These early public pensions were introduced as part of an efficient contract necessary to attract workers in an era when employees often stayed with the same employer for life (Clark, Craig, and Wilson 2003, chap. 2, 154-55). They originated as disability plans and often were fully funded by the employee (Clark, Craig, and Wilson 2003,167). These plans were offered by elected officials who recognized that government jobs were increasingly becoming lifetime careers rather than temporary appointments tied to a particular policy maker's or party member's term in office. This created a political opportunity for individuals running for office to secure electoral support in exchange for increased pension benefits, especially when these pensions were overfunded during periods of strong economic growth. Investment risk was initially minimized by restricting investment to the safest categories, but these restrictions were quickly subverted by governments that used public-pension assets to support their own debt or channeled them as investments to politically connected banks (Clark, Craig, and Wilson 2003, 205, 222).

The federal government followed suit in 1920 and adopted a universal pension plan for federal employees. (4) States began to experiment with retirement assistance to the elderly in need, a program that was eventually subsumed and expanded at the federal level to all retirees in the form of Social Security, largely due to the growing political...