- Pennsylvania’s unfunded pension liabilities are large with only 65% funded based on Dec. 2011 figures.
- If Pennsylvania’s public pensions had to use the same standards as private sector pensions, less than 40% would be funded.
- Bottom line: Pennsylvania, among other states, faces large unfunded pension liabilities due to the structure of DB plans.
Download PDF Chairman Metcalfe, Chairman Benninghoff, and Members of the Committees:
Thank you for the opportunity to testify with regard to public pension reform in the Commonwealth of Pennsylvania. Before elected officials enact reforms, they should have a solid understanding of how current public employee pension plans work and what those plans' true liabilities are. There is significant misunderstanding throughout the country in this regard. For that reason, my testimony consists of a number of broad points I believe are important for policymakers to digest prior to considering broader pension reforms.
Pennsylvania's unfunded pension liabilities are large. Based on December 2011 figures, the Pennsylvania State Employees Retirement System (SERS) is 65 percent funded and faces unfunded liabilities of around $14.7 billion. Worse, the funding ratio is projected to decline to only 55 percent by 2015 before recovering to 63 percent by 2022.
"If Pennsylvania's public pensions were required to use the same accounting standards as private sector pensions, Pennsylvania SERS total unfunded liabilities would be approximately $42 billion and the plan would be less than 40 percent funded." Pension shortfalls are much larger using accurate accounting. If Pennsylvania's public pensions were required to use the same accounting standards as private sector pensions, Pennsylvania SERS total unfunded liabilities would be approximately $42 billion and the plan would be less than 40 percent funded.1 The discrepancy stems from the fact that public pensions may "discount" their liabilities using the high interest rates they project to earn on their investment; for Pennsylvania this rate is 7.5 percent. The problem is that pension benefits are guaranteed while pension investments - in stocks, hedge funds, private equity and the like - can be highly risky. Private sector defined-benefit (DB) pensions must discount their liabilities using lower interest rates to reflect the low risk of the benefits, regardless of how much risk pensions take on the investment side. Professional economists almost universally believe that public pensions should discount their liabilities using low interest rates. Robert Merton, the 1997 winner of the Nobel Prize in Economics, writes that "The conventional actuarial practice of using the expected return on the assets of the pension fund to determine the discount rate for valuing pension liabilities systematically understates their value by large amounts."2
Public pension accounting is unique. Only U.S. public pensions may discount guaranteed benefit liabilities using the high rate of return on a risky investment portfolio. In private sector pensions, public pensions in other countries, economic theory, and the day-to-day practice of financial markets, the discount rate applied to a liability is a function of the risk of the liability, not of any assets used to fund the liability. As Donald Kohn, then Vice-Chairman of the Federal Reserve Board, put it in 2008, "The only appropriate way to calculate the present value of a very-low-risk liability is to use a very-low-risk discount rate."3 Other government agencies, such as the Congressional Budget Office, have written in favor of using lower discount rates to match the low risk of pension benefits.4 The bond ratings agency Moody's recently announced that it will no longer accept pension liabilities calculated using high discount rates. Moody's instead will calculate pension liabilities itself using the much lower yield paid on high-quality corporate bonds. Nationally, this switch will roughly triple unfunded pension liabilities.5
Taking more investment risk won't reduce pension liabilities. A public pension legally must pay benefits even if its investments fail to achieve their forecasted returns. Thus, there is a contingent liability on the government - and on taxpayers - to make good if the plan's assets fall short. Investing in riskier assets raises the plan's expected investment returns and lowers current required contributions. But it also increases the contingent liability placed on future taxpayers, which rises along with the risk of the investments. Both economic theory and market prices for financial instruments that insure against investments falling short of a stated goal (known as "put options") show that the total value of the liability stays the same regardless of how the plan invests. Moreover, this total value can be calculated by discounting pension liabilities at a low-risk discount rate.6
Neither will switching to defined contribution (DC) pensions. A pension's unfunded liabilities reflect the difference between the benefits it currently owes and the assets it currently holds. Shifting a pension plan from a defined benefit (DB) to a defined contribution (DC) structure won't reduce the benefits owed under the existing DB plan. Regardless of how policymakers choose to reform pensions, they must plan for ways to pay benefits that are currently owed.
But DC pensions will keep unfunded liabilities from growing. Using accurate accounting, unfunded pension liabilities are large and continue to grow. That is to say, plans aren't even fully funding the new benefits earned each year. A DC pension, however, is always fully funded because the employer's obligation is only to make a contribution to employees' accounts each year, not to guarantee a fixed benefit years or decades in the future. Unlike DB pensions, whose assumptions can be tweaked in a number of ways to make liabilities appear smaller, DC pensions are transparent to the government, to public employees and to taxpayers.
There aren't any "transition costs" associated with shifting to DC pensions. Some claim that when a government implements a DC pension for its employees that it incurs significant "transition costs" due to a requirement to speed up the payment (or "amortization") of unfunded pension liabilities from prior years. This is not true. If a DC pension is created as a new tier within the existing plan, there is not even an accounting requirement to speed up amortization. Moreover, accounting rules apply only to pension disclosures; state and local governments are free to set their own funding policies, a point that the Governmental Accounting Standards Board has stressed. A number of states have established DC pensions without incurring any transition costs, so these should not be seen as an impediment to reform.7
Public pension benefits are more generous than private plans. In 2011, a Pennsylvania state employee who retired after at least 30 years of service received a benefit equal to 70 percent of his final salary.8 Adding in Social Security, he would receive as much in retirement as when he was working. Very few private sector workers with 401(k) plans will receive benefits anywhere near that level of generosity. Overall, Pennsylvania state government employees receive one of the most generous benefit packages of state workers anywhere in the nation, enough to push combined pay and benefits above private sector levels.9 Reduced pension benefits would not cause Pennsylvania state employees to be less well-paid than private sector workers with similar levels of education or experience.
Failure to make pension contributions is equivalent to borrowing-at a very high rate. When a sponsoring government fails to make its full pension contributions, it is not released from its obligation to pay full benefits. Instead, contributions must be made up in future years, plus interest at whatever rate of return the plan assumes for its own investments. In Pennsylvania's case, this is mathematically equivalent to the government borrowing at a 7.5 percent interest rate at a time when explicit government debt carries much lower yields.
Bottom line. Pennsylvania, along with other states, faces large unfunded pension liabilities due to the structure of DB plans. The gravity of the problem is obscured by specialized accounting rules applied to these pensions. These discrepancies could be addressed by switching to accounting practices followed by the private sector and the rest of the world. While shifting to DC pension plans would not eliminate existing unfunded liabilities, it would ensure that new unfunded liabilities are not generated while making pension generosity more comparable to private sector plans.
Andrew Biggs is a resident scholar at AEI.
1 This figure assumes a 4 percent discount rate.
2 Robert C. Merton. Introduction to Pension Finance, by M. Barton Waring. Wiley Finance. 2012.
3 Kohn, Donald L., "Statement at the National Conference on Public Employee Retirement Systems Annual Conference." New Orleans, Louisiana, May 20, 2008.
4 Congressional Budget Office. "The Underfunding of State and Local Pension Plans." May, 2011. 5 Moody's Investor Services. "Request for Comment: Adjustments to US State and Local Government Reported Pension Data." July 2, 2012.
6 See Biggs, Andrew G. "An Options Pricing Method for Calculating the Market Price of Public Sector Pension Liabilities." Public Budgeting and Finance, Fall 2011.
7 See Robert Costrell. "GASB Won't Let Me": A False Objection to Pension Reform." Policy Perspective Published by the Laura and John Arnold Foundation. May 2012. 8 See Pennsylvania State Employees' Retirement System, 2011 Comprehensive Annual Financial Report, p.87.
9 Author's calculations based on data from the Current Population Survey, the National Compensation Survey and pension and retiree health actuarial disclosures.