Pension Security Bonds
Pension Security Bonds
Abstract and Keywords
This chapter argues that the federal government should be concerned about state pension liabilities. In the absence of fundamental reform, some large state pension funds may not last through the decade. When the funds run through their assets, the size of promised benefit payments will be so large that raising state taxes enough to make these pension payments will be infeasible. The federal government will therefore face massive and likely irresistible pressure to bail out the affected state governments. The chapter proposes reforming states in pension trouble by closing defined benefit pension plans to new workers, an arrangement called a “soft freeze.” Current employees would continue to earn traditional pension benefits under the existing programs, but retirement benefits for future workers would come under a new defined contribution plan. Once a state or municipal government has stopped unfunded pension liabilities from growing, it becomes more feasible, and less costly, for the state to issue debt to improve its liquidity situation.
the federal government should be worried about state pension liabilities. In the absence of fundamental reform, some large state pension funds may not last through this decade. When the funds run through their assets, the size of promised benefit payments will be so large that raising state taxes enough to make these pension payments will be infeasible. Just as the European Union is not standing back to watch Greece fail, the federal government will face massive and likely irresistible pressure to bail out the affected state governments.
Take Illinois, for example. Even if its main three pension funds earn 8 percent returns and the state makes enough contributions to secure new benefits in the coming years, those funds will run out of money in 2018. At that time, benefit payments owed to the workers who are already in today’s state workforce will be an estimated (p.202) $14 billion per year. That is half of the $28 billion in total general fund revenues that Illinois is expected to have received in 2010.
While this problem is particularly severe in states such as Illinois, Connecticut, and New Jersey, many more states have public pension systems that appear unsustainable even on a 15-year horizon (see Rauh 2010). The situation will be exacerbated if mobile taxpayers, frustrated with a large portion of their tax remittances going to pay pensioners, flee to states that can provide lower taxes and higher levels of ser vice. The total size of the potential federal pension bailouts would likely exceed the recent bailout of the U.S. financial system.
When a country gets into fiscal trouble, the policy prescription is usually that it should implement fiscal austerity measures such as tax increases, spending cuts, and pension reforms. These measures essentially bring future bud gets closer to balance and improve the country’s solvency. The tighter bud get also allows the country to borrow the large sums of money that will prevent the legacy liabilities from shutting down the government. The borrowing provides the country with the liquidity to keep operating.
In the case of the U.S. states, the legacy liabilities are the defined benefit (DB) pension promises to state and local government workers. Under a number of state constitutions, promised pension benefits have a legal priority that is senior even to general obligation bonds. States cannot simply back out of these payments. Troubled states that are interested in preventing a crisis should therefore take a page out of the book on sovereign fiscal crises. They should combine fiscal and pension reforms to improve solvency with increased borrowing to keep operating through the flood of benefit payments coming over the coming decades.
A key reform for states in pension trouble would be the closing of DB pension plans to new workers, an arrangement called a “soft freeze.” Current employees would continue to earn traditional pension benefits under the existing programs, but retirement benefits (p.203) for future workers would come under a new defined contribution (DC) plan. The Michigan State Employees Retirement System enacted such a policy for its new workers starting in 1998, although Michigan’s systems for public school employees, state police, and judges still are exclusively DB and have about 460,000 members combined.
These changes must be enacted with an eye toward giving new public employees adequate retirement benefits. Those who are not in Social Security must be brought into Social Security, and the new DC plan must be adequate for supporting the retirement of public employees. The new plan must have automatic enrollment, matching employer contributions, low fees, good investment choices, sensible default allocations, and reasonably priced offers for annuities at retirement.
Once a state or municipal government has stopped unfunded pension liabilities from growing, it becomes more feasible, and less costly, for the state to issue debt to improve its liquidity situation.
A Federal Role
If states were acting in our national interests, and indeed in their own long-term interests, they would be considering these actions themselves. Unfortunately, to the extent that state governments have taken action, it has been largely with cosmetic adjustments to existing systems. It therefore falls to the federal government to give states incentives for pension reform, so that U.S. taxpayers do not ultimately bear the burden of bailing out profligate states.
One area where incentives could be offered is through the treatment of bonds used to fund pensions. Under current law, bonds floated by states to fund pensions are fully taxable. As a result, issuing debt to fund pension plans is considerably more expensive than issuing regular tax-exempt municipal bonds, or the federally subsidized (p.204) Build America Bonds, on which the federal government reimburses 35 percent of all coupon payments directly to the state.
We propose that the federal government cut a deal with the states. A state should be allowed to issue tax-subsidized bonds for the purpose of pension funding for the next 15 years—if and only if the state government agrees to take three specific measures to stop the growth of unfunded liabilities:
1. The state must close its DB plans to new employees and agree not to start any new DB plans for at least 30 years;
2. The state must annually make its actuarially required contribution (ARC) left over from the existing DB plans; and
3. The state must include its new workers in Social Security and provide them with an adequate DC plan, again for at least 30 years. To this end, the federal government should start a Thrift Savings Program for state workers and operate it alongside the existing Thrift Savings Program for federal workers.
The tax subsidies for these new Pension Security Bonds would work like Build America Bonds, with the federal government paying 35 percent of coupon payments directly to the state. Only the amount of the ARC will be tax deductible, to prevent states from overfunding plans and to bring the bonds to the market gradually over 15 years.
How Much will it Cost?
State ARCs for fiscal year 2008 were approximately $67 billion, so the federal government is allowing potentially this much of new tax-exempt borrowing for each of the next 15 years. Assuming that the growth of the ARC is no greater than the discount rate that should be used to discount them, the plan would bring $1.0 trillion (= $67 billion × 15 years) in debt to the market. There is some risk of higher ARC growth in the first several years of the plan, but the fact that fewer and fewer workers will be on the DB plans will limit the ARC growth.
(p.205) If states issue 30-year bonds, the tax subsidy would cost around $250 billion. However, a large fraction of the costs will be offset by the fact that new state workers would be in Social Security. According to estimates by Diamond and Orszag (2005), if all newly hired state and local government workers were on Social Security, it would eliminate 10 percent of the program’s 75-year actuarial deficit, which today stands at $5.3 trillion. A little over two-thirds of state and local workers are on state-sponsored DB pension plans, so bringing these workers into Social Security would save 7 percent of that deficit, or $370 billion. If guarantees by the state to bring new workers into Social Security are only valid for half of the 75 years, the savings from the Social Security expansion would still be over $175 billion. These savings would be borne to some extent by state governments and future state employees paying into Social Security.
Including these gains to the Social Security system, the net cost to the federal government of the entire stabilization program would be only $75 billion. It would prevent a trillion-dollar crisis in less than a decade.
What about the cash flow of the states themselves? Social Security costs 12.4 percent of pay, divided equally among employer and employee. A typical DC plan costs a total of around 9 percent of pay (Poterba, Rauh, Venti, and Wise 2007). Total costs would therefore be 21.4 percent of pay. As a comparison, the plans offered to California state employees cost anywhere between 21.6 percent of pay for the least expensive workers to 40.1 percent of pay for the Highway Patrol Plan. Our proposal would allow states to borrow to meet their share of those costs for existing workers while establishing more sustainable systems for new workers. States would be paying DC contributions and the Social Security share only for new workers, so initially those cash flows would in fact be quite small, given that new hires are a small proportion of the workforce each year.
It is important to take measures that prevent the incoming revenue from new workers into Social Security from crowding out other (p.206) adjustments that the federal government would otherwise have made to Social Security. Separate accounting should be made of the contributions from new state workers in order to discourage the federal government from raiding this pile of money to avoid other fixes it could have made to the program.
Secure Existing Promises and Stop Unfunded Liabilities
This plan offers substantial benefits to numerous parties. Specifically, pension promises made by states to their existing workers become more secure, since the funding of these obligations will greatly improve. New state workers get a retirement plan that is more than an empty promise. Taxpayers avoid massive future tax increases and loss of public services. And, critically, state politicians will no longer be able to use pensions as a vehicle for borrowing off the books at horizons that extend beyond their political careers.
While DC pension plans have some drawbacks, they are immune to the key accounting and accountability issues that have brought many states to the brink of insolvency. States and their employees have for decades been operating under a pretense that promised pension benefits do not represent a real cost, since they will not have to be met for many years. Our plan shifts $75 billion of pension costs for existing workers onto the federal government, but it would prevent a much larger future bailout.
It is instructive to compare our solution to a more extreme idea of waiting until funds run dry and then attempting to cut benefits. This approach would lead to shutdowns of state governments while legal battles are waged over employee rights. Many attempts to cut benefits would fail. Constitutional protections are strong in some of the more troubled states. Municipal bankruptcies such as Orange County in 1994 and the ongoing situation in Vallejo have shown that it is very (p.207) difficult to cut pension benefits—even when bondholders and trade creditors are being impaired.
Some might argue that the federal government should simply take a hard line and announce that it will not, under any circumstance, help states with their pension problems. This approach suffers from the classic problem of time inconsistency. The inability to make credible commitments against bailouts made it impossible for the U.S. government to refuse to bail out large banks. There isn’t even a bankruptcy procedure in place for failing states to restructure their debts, as there is for corporations (Chapter 11) or municipalities (Chapter 9).
Indeed, the federal government would almost certainly come to the rescue rather than watching a state fail. It is therefore imperative that we act today by giving states incentives to put themselves back on a path to fiscal sustainability.
References and Further Reading
Diamond, Peter A., and Peter R. Orszag. 2005. “Saving Social Security: The Diamond-Orszag Plan.” The Economists’ Voice 2, no. 1: art. 8. Available at www.bepress.com/ev/vol2/iss1/art8.
Poterba, James, Joshua Rauh, Steven Venti, and David Wise. 2007. “Defined Contribution Plans, Defined Benefit Plans, and the Accumulation of Retirement Wealth.” Journal of Public Economics 91, no. 10: 2062–86. Available at http://econ-www.mit.edu/files/428.
Rauh, Joshua. 2010. “Are State Public Pensions Sustainable? Why the Federal Government Should Worry About State Pension Liabilities.” National Tax Journal, May 15. Available at http://ssrn.com/abstract=1596679.
Joshua Rauh is Associate Professor of Finance at the Kellogg School of Management at Northwestern University and NBER. Robert Novy-Marx is Assistant Professor of Finance at the University of Chicago Booth School of Business and NBER.