Manhattan Institute for Policy Research.
Subscribe   Subscribe   MI on Facebook Find us on Twitter Find us on Instagram      


No. 61 April 2010


It's Worse Than You Think

Josh Barro, Walter B. Wriston Fellow, Manhattan Institute for Policy Research

Stuart Buck, Distinguished Doctoral Fellow, Department of Education Reform,University of Arkansas

This report was co-sponsored by the Foundation for Educational Choice

Executive Summary

Center for Civic Innovation.


CR 61 (PDF)

Steve Malanga talks with Josh Barro and Stuart Buck about the next great unfunded liability.
The Teacher pension Nightmare, Josh Barro,
Josh Barro on Fox Business's "Varney & Co." 4-19-10
Josh Barro on NY1 News 4-14-10
Josh Barro on Talk Radio Network's "The Jerry Doyle Show" 4-22-10
Stuart Buck on WVON's "The Other Side with Charles Butler" 4-22-10
Josh Barro on WBAL's "The Ron Smith Show" 4-20-10
Bloomberg Radio's "The Hays Advantage" 4-15-10
XM Radio's "The Peter Cezbs Show" 4-15-10
Teacher Pension Alarm, The Journal Gazette editorial, 4-24-10
Rockwood becoming financially sound, but facility problems remain, Our Town, 4-25-10
Report says Indiana teacher retirement fund far short of assets to pay teacher pensions, Associated Press, 4-21-10
Teacher Pension Plans Could Trigger Next Big Bailout Crisis, Peter Roff, Thomas Jefferson Street blog on U.S. News & World Report, 4-21-10
California Pension Plan Would Vest 10 Years Later, BusinessWeek, 4-21-10
Report: Indiana short on funding teacher pensions, Courier-Journal, 4-21-10
Report says Indiana teacher retirement fund far short of assets to pay teacher pensions, Canadian Business, 4-22-10
Study: Teacher Pension Plans on Shaky Ground Hawaii Reporter, 4-15-10
Two sides disagree on teachers' pensions, Atlanta Journal-Constitution, 4-18-10
Connellsville Area to absorb additional pension costs into budget, Daily Courier, 4-19-10
Lawmakers question report on teacher pension fund, Charleston Daily Gazette, 4-14-10
Analysis: Pension funds for teachers are short by billions of dollars, Newark Advocate, 4-15-10
Study finds teacher pension funding gap, but state disagrees, Atlanta Journal Constitution, 4-14-10
Analysis says pension funds for teachers are short by billions, Daily Record, 4-14-10
State teacher pension plans face nearly $1 trillion in unfunded liabilities, Washington Examiner, 4-14-10
Report: Teacher Pensions Face $36 Billion Shortfall, NY1 News, 4-14-10
Report: Teacher pension fund nearly $1 billion in the red, FOX 59 WXIN, 4-14-10
A fiscal black hole grows deeper, Atlanta Journal Constitution, 4-14-10
Report Disputes Officials Estimate of Pension Funds, WTAQ News Talk, 4-14-10
New Report on Teacher Pensions, Stuart Buck responds to Ed Sector blog "Quick and the Ed" on Jay P. Greene's Blog, 4-14-10
UPDATE: West Virginia Has Worst Funded Teacher Pension Plan in Nation, West Virginia Watchdog, 4-14-10
Public and Private Pensions, Ed Sector blog "Quick and the Ed" 4-13-10
Study: Taxpayers owe teacher pensions $933B, Crain's NY Business, 4-13-10
Teachers' pension fund under-funded by $36B, New York Post, 4-13-10
Teacher Pension Deficit at $900 Billion May Be Triple Reported, BusinessWeek, 4-13-10
Analysis: Teacher Pensions Underfunded by $900 Billion, Education Week, 4-13-10
Teacher Pension Deficit at $900 Billion May Be Triple Reported, San Francisco Chronicle, 4-13-10
W.Va. teacher pensions most underfunded, report finds, Charleston Gazette, 4-13-10
Teacher Pension Plans in Kansas Among Worst Funded, Kansas Watchdog, 4-13-10
Wisconsin pension funding for teachers falls $10.9 billion short, report says, Milwaukee Journal Sentinel, 4-13-10
Pension crisis could be even worse in Pennsylvania, Delaware County Daily Times, 4-14-10
Academics point to $900bn-plus hole in US teacher funds, Financial News, 4-14-10
Report: Pension funding worrisome, Washington Observer Reporter, 4-14-10
Analysis says pension funds for teachers are short by billions; Gap isn't all because of the market crash, USA Today, 4-14-10
Report: Teacher Pension Plans Shaky, Pittsburgh Post-Gazette, 4-14-10

Teacher pension funding gaps are three times greater than what states report, states a new Manhattan Institute/Foundation for Educational Choice study. Authors Josh Barro and Stuart Buck reveal the major disparity between what states report and the true value of unfunded liabilities for teacher pensions. These liabilities for all 50 states now total almost $1 trillion—an unfunded burden that states must pay over time at taxpayer expense.

Table of Contents:
Executive Summary
About the Authors
I. Background
II. Our Calculation of State Governments' True Liabilities For Teacher Pensions
III. Our Recommendations

To all the other fiscal travails facing this country’s states and largest cities, now add their pension obligations, which are far greater than they may realize or are willing to admit. This paper focuses on the crisis in funding teachers’ pensions, because education is often the largest program area in state budgets, making it an obvious target for cuts.

Although it is generally acknowledged that education is the foundation of every modern society’s future prosperity, schools unfortunately will have to compete with retirees for scarce dollars. This competition is uneven, because retirees have a legal claim on promised pension benefits that supersedes schools’ budgetary needs. Consequently, Americans can look forward to higher taxes and cuts in services, resulting in fewer teachers, bigger classes, and facilities that are allowed to deteriorate. In several states, these developments have already arrived.

The crux of the problem is the gap between assets and liabilities affecting the fifty-nine pension funds that cover most public school teachers in America. Some of these are general state-employee pension funds, while others cover only teachers. Among the findings of our study of these funds:

  • All fifty-nine pension funds studied face shortfalls.
  • California, the most populous state, has the largest unfunded teacher pension liability: almost $100 billion.
  • The worst-funded plan in our sample is West Virginia’s, which we estimate to be only 31 percent funded.
  • Five plans are 75 percent funded or better: teacher-dedicated plans in the District of Columbia, New York State and Washington State and state employee retirement systems in North Carolina and Tennessee that include teachers.

The general picture is not a good one. According to the fifty-nine funds’ own financial statements:

  • Total unfunded liabilities to teachers—i.e., the gap between existing plan assets and the present value of benefits accrued by plan participants—are $332 billion.
  • According to our more conservative calculations:

    • These plans’ unfunded liabilities total about $933 billion.

    In addition, we have found that:

  • Only $116 billion, or less than one quarter, of this $600 billion discrepancy is attributable to the stock market drop precipitated by the 2007 financial crisis.
  • The Dow Jones Industrial Average would have to nearly double overnight to make up for the present underfunding of these plans.
  • What explains the rest of the gap between the funds’ estimates and our own? The funds aggressively “discount” the cost of paying benefits in the future because they assume that stocks’ values will be much higher by the time the funds have to pay out those benefits. This assumption permits public officials to contribute fewer dollars toward satisfying these plans’ obligations, and thus to avoid taking the cautious but unpopular step of raising taxes or cutting services.

    Under current guidances, which are prepared by separate bodies, state pension funds are able to set aside fewer assets than their counterparts in private companies to cover equal liabilities. Private pension plans may invest in stocks and other higher-risk assets, but those plans may not reduce their pension funding on the basis of the superior performance expected of these types of assets. This is because those higher returns are accompanied by greater risk that returns will fall short of expectations. Yet pension funds’ obligation to retirees present and future does not diminish accordingly.

    By contrast, public pension plans are permitted to base the amount of money they need today to meet their future obligations on the higher expected performance of stocks, allowing sponsors to cut their contribution rates and hope the markets perform as anticipated. Unfortunately, markets can drop instead of rising, as they did in 2008-2009, when the large decline in the Standard & Poor’s 500 index of stocks created especially severe shortfalls among public pension funds. With formal bond indebtedness of U.S. states and localities reaching approximately $2.4 trillion, the shortfalls in teacher pensions alone increased the indebtedness of state and local governments by roughly one-third.

    The purpose of this paper is not to instruct pension funds in how to invest their funds. Rather, it is to recommend that they account for their liabilities in the manner of private pension plans, which must estimate the present cost of future liabilities on the basis of the lower returns that high-quality corporate bonds pay. They do so because the likelihood that these instruments will fail to make payments to their owners, the pension funds, is about as great as the risk that retirees will not receive their promised payments in the promised amounts. Using these accounting rules, public plans (like private ones) would be denied the opportunity to short-change their pension plans by assuming strong asset performance.

    Unfortunately, accounting reforms will not eliminate the accrued liability, which represents income already earned by public employees but not yet paid. States must simply amortize these costs over time, at taxpayer expense. In part they have grown so large because elected officials have not been held accountable for them: while the cost of higher retirement benefits is off in the future, the cost of higher wages is in the present, and thus visible and felt. Visible or not, the promise of future pensions is a very real cost of hiring teachers, one just as real as the teachers’ current salaries.

    States can start by accounting honestly for the current costs of future benefits. If they did so, they would reduce the temptation of their elected officials to be overly generous in awarding benefits. Going forward, there are structural changes they can take that would avoid funding shortfalls and rein in out-of-control public pensions:

    • States should consider shifting to defined-contribution retirement plans, especially on behalf of new and young employees; this is the norm in the private sector and was adopted successfully by Michigan in the 1990s. States are not obligated under such plans to provide any particular level of benefit.
    • In cases where defined-contribution plans face major political resistance, states should consider hybrid options like cash-balance plans and TIAA-CREF, the latter having provided a version of defined-contribution retirement saving for employees of public colleges and universities for decades.

    About the Authors

    Josh Barro is the Walter B. Wriston Fellow at the Manhattan Institute focusing on state and local fiscal policy. He is the co-author of the Empire Center for New York State Policy’s “Blueprint for a Better Budget.” He writes weekly on fiscal issues for and has also written for publications including the New York Post, Investor’s Business Daily, the Washington Examiner, City Journal, and His commentary has been featured on CNN, Fox News Channel, CNBC, the Fox Business Network, and Bloomberg Television.

    Prior to joining the Manhattan Institute, Barro served as a staff economist at the Tax Foundation, where he wrote the 2009 “Tax Freedom Day” report. He also wrote several critical analyses of state tax and budget proposals and gave testimony on tax reform before officials in Arkansas, Louisiana, Maryland, New Jersey, Rhode Island, and South Carolina. Previously, he worked as a commercial real estate finance analyst for Wells Fargo Bank. Barro holds a B.A. from Harvard College.

    Stuart Buck is a Ph.D. candidate in the Department of Education Reform at the University of Arkansas. He is the author of a forthcoming book, Acting White: An Ironic Effect of Desegregation, from Yale University Press (2010). He has also authored numerous scholarly articles in journals such as the Review of Public Personnel Administration, Harvard Law Review, Harvard Environmental Law Review, Stanford Technology Law Journal, Administrative Law Review, Federal Communications Law Journal, and the Case Western Law Review. Buck received a J.D. with honors from Harvard Law School in 2000, where he was an editor of Harvard Law Review; he thereafter clerked for Judge David Nelson of the Sixth U.S. Circuit Court of Appeals and Judge Stephen F. Williams of the U.S. Circuit Court of Appeals for the D.C. Circuit.



    Since the last quarter of 2007, the start of the most recent recession, nearly every state has experienced significant financial distress. This year, forty-eight states faced budget shortfalls ranging from 2 percent to 49 percent, with a weighted average of 17 percent.[1] California’s fiscal situation was so desperate in 2009 that, for a time, it was forced to issue IOUs to employees and vendors in place of cash payments. Despite the help that federal stimulus money provided in bridging state budget gaps, nearly every state had to resort to spending cuts, tax increases, or both to achieve balance. And revenue is unlikely to improve enough by the time the stimulus expires, in 2011, to prevent more fiscal pain. As the National Association of State Budget Officers itself acknowledged in its December 2009 report, “Fiscal conditions significantly deteriorated for states during fiscal 2009, with the trend expected to continue through fiscal 2010 and even into 2011 and 2012” (NASBO 2009, p. vii).

    Education, as the largest or next-to-largest program area in most state budgets, is an obvious target for budget cuts. Meanwhile, the cost of education, including pay increases in multiyear teacher contracts, continues to rise. These budget pressures will inevitably force localities either to raise property taxes or to reduce education spending. In March, for example, the governor of Illinois threatened to cut aid to school districts by $1.3 billion, unless the legislature raised the income tax by 1 percent.

    An ominous backdrop to schools’ current funding difficulties is the underfunding of teacher pension plans, which would have to draw on the same pool of funds as current school operations if they were to return to solvency. Indeed, Illinois is facing annual contributions of over $4 billion that it will have to make to pension plans for public employees, including teachers. This figure is four times what it was a decade ago.

    State and local governments collectively admit to underfunding teacher pension plans to the extent of some $332 billion, according to figures in their Comprehensive Annual Financial Reports. But that estimate is far too low. In order to produce such estimates, state governments are assuming, on average, that their investments will appreciate at about 8 percent per year for an indefinite period. Then, they use this 8 percent figure to discount future pension obligations to a present value, which is the estimate of funds that must be set aside currently to pay all future obligations.

    Although this method of calculation is prescribed by the Government Accounting Standards Board, it is not a reasonable way to estimate out-year pension liabilities, even if 8 percent were a reasonable estimate of plan assets’ rate of return. This is because states are not able to pass along any of the risks associated with these higher returns to plan beneficiaries. Such shortfalls would have to be borne by the obligor, not the beneficiary, as would be the case in a defined-contribution plan.[2]

    Unlike public plans, private-sector pensions are required to use discount rates close to the yield on high-quality corporate bonds, which make regular payments in fixed amounts with a low degree of risk to the bondholder. Currently, the rates paid by such bonds are about 6 percent, which translates into a discount rate in the 6 percent range. Pension-plan assets may be invested in higher-return assets, but because the added risk they pose cannot be passed on to plan beneficiaries (except, in certain cases, in a corporate bankruptcy), the plans may not increase their discount rates to reflect higher expected returns.

    If the same standards that govern private-sector pension plans are used and current stock-market values are adjusted to reflect today’s stock-market values, actual financial liabilities for public school teacher pension plans are approximately $933 billion, which is close to triple the official estimate.

    This actuarial deficit is not primarily the result of the recent stock-market decline. Of the approximately $600 billion by which unfunded liabilities have been underestimated, in our view, only $116 billion is due to the stock-market decline of the last two years. The lion’s share of the underestimation ($484 billion) stems from the funds’ use of much too aggressive discount rates for future liabilities.

    For the states, the most obvious answer to the underfunding problem is drastically increasing their annual contributions, but beleaguered taxpayers, already struggling to provide for their own retirements, may balk at suffering steep cuts in state services or paying higher taxes so that state employees can retire comfortably. The Associated Press recently quoted a local resident on the underfunded status of state pensions: “‘I think it’s ridiculous,’ said Larry Rausch, 71, of Lancaster, who retired from a sales job at Sears in 1998, before the owner of Kmart purchased the retailer and slashed retirement benefits. ‘I don’t know how they can expect guys like me to pay their retirement.’”[3] And the president of the Maryland Senate, facing a $2 billion deficit in his state, has said that the cost of teacher pensions, which was approaching $1 billion a year, was “not sustainable.”[4] While Ohio and Maryland have teacher pension funding problems that are worse than average (of the fifty-nine systems in our study, Ohio has the twenty-fourth worst funding ratio and Maryland the twenty-eighth), they are not outliers; many states face problems of similar magnitude.

    Fully funding state teacher pensions would dramatically affect states’ and localities’ ability to fund other important commitments. For example, an Ohio newspaper found that fixing the teachers’ retirement plan in Cleveland alone “could cost more than the Cleveland schools spend on textbooks each year and more than it spends to get students to school. The budget hit to Cleveland schools would be about what it costs to put 110 teachers in classrooms.”[5] Taxpayers need to realize that lavish teacher pensions come at the cost of denying funding to other important educational objectives.

    I. Background


    "Nearly 20 million employees and 7 million retirees and dependents of state and local governments—including school teachers, police, firefighters, and other public servants—are promised pensions” in their retirement (GAO 2008, p. 1). State pensions are tracked by the Public Fund Survey, an annual report compiled by the National Association of State Retirement Administrators and the National Council on Teacher Retirement, which contains statistics on major pension plans covering state and local government employees, including teachers.

    In the Public Fund Survey’s database, there are fifty-nine state and local pension plans that cover teachers. The pension funds that we analyze in this report include more than 9 million active employees and 4 million retirees and manage over $1.5 trillion in assets. In several states, such as Arizona, Florida, and New Hampshire, public school teachers are included in statewide public-employee pension plans. Where possible, we prorate the liabilities of such plans according to the percentage of state employees who work for school districts.[6]

    Every state administers a defined-benefit pension plan in which public school teachers participate, although Alaska’s plan is closed to new participants. Michigan has done the same with its main employee retirement plan, directing new state employees into 401(k)-style retirement plans, but teachers retain their separate defined-benefit pension plan (see Pew 2006, p. 33). Under a typical defined-benefit plan, teachers are promised a specific amount of money regardless of how well or how poorly the plan’s investments have done or any budgetary problems that the plan or the state might be experiencing. Generally, the plans base benefit levels on number of years of service, as well as income in the last few years preceding retirement. For the latter reason, benefits may be inflated by teachers who engineer a brief boost of income at the end of their careers—by taking on a department chairmanship or teaching in a summer program, for example.

    Most states are legally bound to protect pension benefits. Several of them, including New York and Illinois, “provide through specific constitutional provisions that state retirement plans cannot be amended in any way that results in a participant receiving a lower retirement benefit than that which would be payable under the plan terms in effect as of the date the employee first became eligible to participate in the plan” (Monahan 2009, p. 6). “Michigan and Hawaii have state constitutional provisions that have been interpreted as protecting pension benefits accrued to date” (Monahan 2009, p. 9). Still other states, including California, have found pension plans to be akin to contracts; thus, “amending state pension plans may violate the contract clauses of both the federal and state constitutions” (Monahan 2009, p. 11). While such plans can theoretically be changed, the California Supreme Court has held that “changes in a pension plan which result in disadvantage to employees should be accompanied by comparable new advantages.”[7] Consequently, many states find themselves in the “untenable position of being unable to amend their pension plans even with respect to future employee service” (Monahan 2009, p. 26).


    The history of state pensions over the past decade is one of irresponsibility, hubris, and lack of foresight. The booming stock market of the 1990s resulted in stronger than expected asset performance, leading to pension-fund surpluses. Instead of setting aside investment gains for future pension payments, state governments started “shortening vesting periods, increasing the multipliers used in determining benefit amounts, decreasing the age at which employees could receive full retirement benefits and shortening the years of service needed to qualify. New York, New Jersey, Illinois, Pennsylvania, Kentucky, California, Colorado and other states increased benefits” (Pew 2006, p. 8).

    Declining returns in the first decade of the new century rendered those gifts difficult to pay for. Indeed, we have now been through a period of over eleven years of essentially zero overall growth in the stock market—the Standard & Poor’s 500 index closed on January 8, 2010, at 1141.69, a level that it first reached in July 1998.

    In response, the New Hampshire pension system, which reported a $1 billion loss in fiscal 2009, was forced to propose that municipal employers increase their pension contributions by an average of 22.7 percent.[8] Likewise, in Alabama, the Board of Education “voted to recommend freezing the state’s contribution to teachers’ health and retirement programs, and to increase from 5 percent to 6 percent the portion of teachers’ salaries deducted for retirement” (Diel 2009).

    Other jurisdictions have taken actions that have only pushed them further behind. New York City, which has the eleventh-largest teacher pension system in the country, has offered teachers several new pension sweeteners over the last decade, such as retirement at age fifty-five with full benefits, a policy of counting per diem payments to salaried teachers in determining the size of their pension benefits, and others.[9] Over the same period, teacher salaries rose 43 percent, increasing final average salaries and therefore the level of retiree pension benefits. These developments have combined with declining asset values to more than quadruple the city’s annual required pension contribution, after adjusting for inflation. As Gotham Schools put it, “Together, the rising salaries and pension sweeteners have created a perfect storm: increasing costs just as the plan’s performance has plummeted in the down market.”

    The need to pour more money into pension funds will force states either to cut back spending in other budget areas—classroom education, Medicaid, transportation, and so on—raise taxes, or both. For now, many states are deferring payments to pension plans in order to alleviate short-term budget stresses. But doing so only deepens the extent of underfunding. In March 2009, New Jersey passed (and Governor Jon Corzine signed) a pension deferral bill that “let local governments defer up to half of their employee pension obligations this year to stave off tax hikes or layoffs” (Rispoli 2009), even though New Jersey was already “$52 billion, or 45 percent, short of what it should have on hand to cover pensions already earned”[10] (Fitch 2009). In 2003, Illinois governor Rod Blagojevich, who left office in 2009 in disgrace, embraced a plan to “issue debt at a cost of 5.1 percent and then earn 8.5 percent or so investing the proceedings [sic].” This turned into “a disaster” when the market dropped last year, leaving Illinois about $60 billion short (Fitch 2009). Massachusetts, too, was seeking ways to avoid funding state pensions fully, as state law required: according to one state senator, “forcing a $900 million spike in state pension funding would be devastating to other spending accounts” (Jourgensen 2009).

    Deferring the inevitable is not a long-term strategy for achieving financial stability. States and cities will have to make up these deferred payments—with interest—in later years.


    Parties obligated to pay an amount at some future date need to know the size of that obligation in today’s dollars, which will tell them how much money to set aside. That sum can be smaller than the principal amount due because it can earn interest until the due date. If, for example, you owe $10,000 in ten years, and your savings account offers an interest rate of 3 percent, you would need to set aside only $7,441 today. In this example, you have assessed your future obligations using a 3 percent “discount rate”—the rate at which the principal due is discounted over a given period of time to produce the loan’s net present value.

    Pension funds likewise rely on discount rates to tell them how they can meet their future financial obligations. In so doing, they follow the lead of the Government Accounting Standards Board (GASB), an organization that establishes financial standards for state and local governments. GASB “operates independently and has no authority to enforce the use of its standards,” but “many state laws require local governments to follow GASB standards, and bond raters do consider whether GASB standards are followed” (GAO 2008, p. 7).

    In its Statement 25, “Financial Reporting for Defined Benefit Pension Plans and Note Disclosure for Defined Contribution Plans,” GASB advises that a discount rate “be based on an estimated long-term investment yield for the plan, with consideration given to the nature and mix of current and expected plan investments.” So long as average returns are sufficient to cover a plan’s benefits, it is fully funded, according to GASB standards, even if the riskiness of its investment choices creates a greater than 99 percent chance of a funding shortfall, which taxpayers would at some point be responsible for repairing.

    It is the view of many actuaries that discount rates should be set at an 8 percent level, more or less, and left there, so that participants needn’t vary their contributions and the plan’s projections can remain relatively consistent over the years (Jones, Murphy, and Zorn 2009). The argument in favor of this discount rate is that the stock market has historically performed at or above it and that this level of performance should continue if measured over a long-enough time span. Since governments, unlike private companies, rarely dissolve, the argument goes, their pension portfolios should have enough time eventually to match the historical rate of return.

    Plans mostly invested in stocks and other equities use the stock market’s higher returns over long periods of time as their rationale for using discount rates in the 8 percent range. As University of Chicago economists Robert Novy-Marx and Joshua Rauh point out, GASB permits underfunded pension plans to increase their liability discount rates, and thus eliminate their funding gap, simply by increasing the risk profile of their asset portfolio.

    Unfortunately, a plan may fail to meet its 8 percent return target for an extended period, or even see asset values fall significantly, as most portfolios did in 2008-09. If such a period should persist long enough, pension reserves can drop to the point where states are forced to close the gap by drastically increasing pension contributions. Governments’ own indefinite existences do not give them the luxury of waiting indefinitely for the market to recover.

    Instead of using expected asset returns to discount liabilities, some financial economists call for a discount rate that reflects “the riskiness of the liabilities, not the assets” (Brown and Wilcox 2009, p. 10; echoed in Waring 2009, p. 19). Their thinking is that public pension plans are providing a benefit that is essentially guaranteed, come what may. But the gains and income on which pension plans rely to provide that benefit are not guaranteed. To eliminate this mismatch, “discount rates should be derived from securities that have as little risk as the liabilities themselves” (Brown and Wilcox 2009, p. 8), the “risk” of these liabilities being that a pension plan would be able to escape its obligations to beneficiaries, which is exceedingly unlikely. The theory underlying this approach is commonly known as the “market value of liability” (MVL).

    Just as GASB oversees public plans, the Financial Accounting Standards Board (FASB) oversees private plans. It directs them to discount their pension liabilities on the basis of the risk profile of pension liabilities, not assets. FASB’s directive rests on the recognition that firms cannot pass the risk associated with higher returns on to plan participants. Paragraph 44A of FASB Statement 87 reads:

    [A]n employer may look to rates of return on high-quality fixed-income investments in determining assumed discount rates. The objective of selecting assumed discount rates using that method is to measure the single amount that, if invested at the measurement date in a portfolio of high-quality debt instruments, would provide the necessary future cash flows to pay the pension benefits when due. Notionally, that single amount, the projected benefit obligation, would equal the current market value of a portfolio of high-quality zero coupon bonds whose maturity dates and amounts would be the same as the timing and amount of the expected future benefit payments.

    Private plans generally choose a discount rate based on a blended average of corporate bonds in the Moody’s Aa rating range, pegged by Mercer Consulting as of February 2010 at 6.06 percent over a fifteen-year plan horizon, the typical period used by public-sector plans.[11] This yield reflects the risks associated with high-quality corporate bonds; nearly risk-free assets such as U.S. Treasury bonds usually pay considerably less. The inclusion of a risk premium reflects the possibility that the sponsoring corporation could go bankrupt and thus default on its pension obligations.[12]

    It is beyond the scope of this paper to tell public pension funds what their asset mix should be; as with private plans, it can be perfectly appropriate for such funds to invest in a pool of assets whose risks exceed those of fund liabilities. However, we do contend that public pension funds should adopt the private pension practice of discounting liabilities on the basis of the risk to plan participants that they will not receive their benefits, not on the basis of the expected returns of the funds’ asset pools.

    GASB is currently considering whether to revise its rules on public pension accounting and, in particular, what discount rate(s) to mandate (GASB 2009, pp. 32–37). We examine the solvency of teacher pension plans, should they be subjected to the same standards that FASB imposes on private pension plans.

    II. Our Calculation of State Governments' True Liabilities For Teacher Pensions


    Most state and local government pension plans discount their future liabilities at approximately an 8 percent rate, which is set to match the expected return on plan assets. However, as discussed in the previous section, this practice is not in line with the accounting standards of private-sector pensions, which recognize that taxpayers are on the hook for pension obligations even when plan assets underperform. Therefore, we adjust the calculations of teacher pension plans’ present-value liabilities on the basis of discount rates resulting from methodologies in wide use in the private sector.

    This is not an innovation on our part. Novy-Marx and Rauh assembled a data set of state pension-plan liabilities and then calculated what the true liabilities of those plans were likely to be if the discount rate were either the fifteen-year Treasury bond rate or a municipal bond rate. In their words, “while the plans appear almost fully funded under government-chosen discount rates, there is a large probability of significant shortfalls in the future.” They go on to say that the cost of fully insuring future taxpayers and plan participants against these shortfalls—that is, the present value of liabilities above what can be covered by plan assets—approaches $2 trillion, or over 80 percent of the value of all outstanding state and municipal bonds in the United States.

    Our analysis draws on Novy-Marx and Rauh’s methodology but treats the typical public pension plan far more generously by adopting the discount rates used by private pension plans that are in compliance with federal law and FASB standards. Like Novy-Marx and Rauh, we assume a fifteen-year duration for pension plans, which means that accrued plan liabilities are to be paid on average in fifteen years. (Accrued liabilities range from payments due this month to existing retirees to payments that will be made decades from now to workers who are young today.)

    Public Funds Survey data include the estimated actuarial liabilities of fifty-nine state and local teacher pension plans, which we then updated with more recent data from each plan’s Comprehensive Annual Financial Report (where available).[13] Following the Novy-Marx/Rauh formula, we adjust the plan’s liability discount rate (typically between 7.75 percent and 8.25 percent) to the Mercer index figure of 6.06 percent.

    The result is a more realistic estimate of state teacher pension liabilities—around $817 billion, or $484 billion more than state governments currently admit. (A complete table of shortfall amounts for each individual pension plan is included as Appendix A.)


    The picture gets even worse. Besides understating their actuarial liabilities (and therefore their unfunded liabilities, which equal the gap between actuarial liabilities and assets), many plans are overstating their asset values. They do this by failing to immediately recognize asset returns that differ from their target return rates; instead, they amortize the deviation in the rate of return over a number of years, most commonly five. Rising stock prices produce a conservative actuarial valuation; however, the significant drop in stock prices over the last two years, because it is not yet reflected in plan financial statements, has produced actuarial asset valuations that exceed the true market value of plan assets.

    The S&P 500 was 24 percent lower on December 31, 2009, than it was on June 30, 2007. Although we do not draw upon unofficial statistics in this analysis, we know that many public pension plans suffered devastating losses. For example, the Pennsylvania teacher pension plan, PSERS, announced in a press release that it lost over $20 billion between June 30, 2007, and October 1, 2009 (PSERS 2009, p. 5).

    We do not have a clear asset valuation figure for most plans that is more recent than their most recent annual financial report, whose reporting dates vary from June 30, 2007, to June 30, 2009. In an effort to approximate these plans’ current asset market value, we have used a two-step process. First, we substitute the “market value” of assets that each plan reported in its most recent financials for the “actuarial value” or “funding value” that is typically used to calculate funding gaps. Second, we adjust the value of the equity portion of the fund’s investment portfolio to reflect the change in value of the S&P 500 since the date of the last financial statement.[14] After making such an adjustment, the difference between assets and liabilities (as calculated in the previous section) comes to a staggering $933 billion, a further increase of $116 billion.

    Of course, fluctuations in the stock market may reduce the underfunding problem or exacerbate it. Currently, weak stock-market performance is a significant contributor to underfunding, but not the predominant one. Even if the stock market rebounded strongly in the next several years, it would need to rapidly reach levels on the order of Dow 25,000 to offset the effects of overly aggressive discount rates.

    III. Our Recommendations

    The insufficiency of assets in state teacher pension funds is massive and unsustainable. In order to maintain their creditworthiness and satisfy liabilities, states need to take the following steps:

    1. Account for and fund the existing liability; and
    2. Reduce the cost of future benefits.

    Achieving step one won't be easy. Existing accrued but unpaid pension liabilities are functionally equivalent to a state’s unsecured debt. (Indeed, under some state constitutions, they hold an even more privileged position than bond debt.) Even if states can default on these obligations, they generally should not: accrued pension liabilities represent compensation earned for past labor, not compensation promised for future labor.

    The $933 billion in unfunded liabilities will simply have to be paid off over time, with the help of higher tax revenues and/or cost savings in other areas of government. States that choose to put off remedying these funding gaps will see them grow only larger over time, and they will do near-term damage to their credit ratings and ability to borrow. Illinois, which has some of the country’s largest unfunded pension liabilities, saw its Moody’s general obligation bond rating downgraded from A1 to A2 last year. Only California has a lower rating. Among the challenges Moody’s cited in downgrading Illinois were the state’s large gaps in funding pensions and other post-employment benefits.

    The good news is that several decent options are available for handling step two. Unfunded pension liabilities have ballooned in part because states have preferred to incur hidden pension costs instead of imposing visible wage costs. Once states recognize how costly their pension dodges have become, they are likely to take action to contain future benefits.

    One option is to shift away from defined-benefit pension schemes and to defined-contribution options such as 401(k) and 403(b) plans. In lieu of a pension plan, governments can make deposits into individual employee accounts, into which employees may also deposit their own funds. Defined-contribution plans cannot, by definition, develop funding deficits: the employer’s role is limited to making deposits in the retirement account. No longer guaranteed a certain level of returns, it is the account holder who decides to invest the asset balance in stocks or bonds or some combination of the two.

    These sorts of plans have generally supplanted defined-benefit pensions in the private sector: while 84 percent of state and local government employees were offered participation in a defined-benefit pension plan as of March 2009, only 21 percent of workers in private industry (and only 16 percent of the nonunion workers among them) were offered participation.[15]

    Michigan successfully implemented a defined-contribution reform in the 1990s, with state employees hired since 1997 shifted to a defined-contribution system; teachers remain in a separate defined-benefit plan. Florida also offers a defined-contribution alternative to its defined-benefit plan.

    Public-employee unions’ opposition to defined-contribution retirement plans is often fierce, making such a shift politically infeasible in many states. However, allowing existing employees to remain in the defined-benefit system can dampen opposition, as it did in Michigan. A number of more moderate, compromise reforms are also available:

    • A shift from defined-benefit to “cash balance” pension plans. Cash balance plans are a sort of hybrid of defined benefit and defined contribution: employees and employers make contributions, and then the state plan guarantees a rate of return on those contributions. Employees are guaranteed not to lose money on plan investments, but a linear relationship is established between contributions and benefits. Nebraska, for example, started offering a cash balance plan as an option in 2003 (Pew 2006, p. 34). While these plans have the advantage of closely tying the size of the benefit to lifetime compensation (as opposed to, say, a participant’s last three years of income), they do not prevent states from adopting overly aggressive investment targets. Nor do such plans shield states from the consequences of market underperformance.
    • A shift to a TIAA-CREF-style hybrid plan. The Teachers Insurance and Annuity Association–College Retirement Equities Fund has been the leading provider of retirement products to college and university employees for nearly a century. TIAA-CREF offers a variety of products, including traditional defined-contribution pension investments such as mutual funds. The difference between TIAA-CREF and a standard 401(k) plan is more a matter of appearance than substance, but TIAA’s emphasis on variable-annuity investment products (which provide a defined-benefit-like certainty of payouts) could help reassure public employees that pension reform will not leave them destitute in retirement. The fact that many professors at state colleges and universities have long been offered TIAA-type plans instead of defined-benefit pensions should also help in making the case that these are not second-rate kinds of plans.
    • An honest accounting for future defined-benefit liabilities. This is something that all states with defined-benefit pension plans should embrace, but especially those that face significant political or constitutional barriers to fundamental pension reform. A key driver of ever-rising retirement benefit costs is their hidden nature: it is easier today to promise retirement benefits that won’t have to be paid out for years than to give pay raises that come out of present revenues. Accounting accurately in the current period for the future costs of promised benefits—particularly by using a more conservative discount rate for future pension liabilities, as this paper advocates—will raise the currently recognized cost of benefits and constrain the political impulse to be overly generous. No matter what path pension reform may take, state governments have no excuse not to admit the true extent of their unfunded liabilities. They must take serious and substantive steps to prevent financial disaster in the future.



    1. Montana and North Dakota, the only states with no budget gaps to close, benefited from relatively strong performance in the agricultural and energy sectors, respectively.
    2. Additionally, the past twelve years of stock performance (the S&P 500 Index is lower today than it was in 1998) suggests that return expectations of about 8 percent may be unreasonable for portfolios carrying significant risk. For this reason, CalPERS, the largest state employee retirement fund in America, is considering reducing its forecasted rate of return from 7.75 percent to as low as 6 percent. See Gina Chon, “Calpers Confronts Cuts to Return Rate,” Wall Street Journal, March 1, 2010, available at
    3. Associated Press, “Ohio taxpayers asked to cover rising pension costs,” 4 Jan. 2010.
    4. “O’Malley asks superintendents to scour for savings,” Baltimore Sun, 20 Oct. 2009.
    5. “Fixing public-employee pension plans costly,” Columbus Dispatch, 3 Jan. 2010.
    6. In a few cases, state pension plans do not record or track the number of teachers or school-district employees at all. In such cases, we estimate that 45 percent of state pension liabilities have been accrued on behalf of teachers, which is the average figure in other states.
    7. Betts v. Bd. of Admin., 21 Cal. 3d 859, 864, 582 P.2d 614 (1978).
    8. New Hampshire Retirement System press release, November 10, 2009, available at
    9. See
    10. To be sure, Corzine was not the first New Jersey governor to engage in such irresponsible behavior. As Forbes noted, “Republican Governor Christine Todd Whitman played the game in the 1990s by using rosy investment-return projections to justify skipping two years of pension contributions. Her successor, Democrat James McGreevey, kept the practice going” (Fitch 2009).
    11. See Goldman Sachs Global Markets Institute, “Accounting Policy Update: Big Contributions to Pension Plans, but Still Underfunded,” September 16, 2009, available at; Peter Fortune, “Pension Accounting and Corporate Earnings: The World According to GAAP,” Federal Reserve Bank of Boston, Public Policy Discussion Papers No. 06-2, p. 15; and Mercer Pension Discount Yield Curve and Index Rates—December, updated January 5, 2010, available at
    12. There are good arguments for subjecting governments to even more conservative pension accounting than private firms must use. Because private firms can go bankrupt and default on their pension obligations, pension benefits are not truly without risk to the beneficiary. Since a state is less likely to default on its pension obligation than a corporation is, especially states whose pension benefits are constitutionally guaranteed, a lower rate is probably called for, increasing the size of the unfunded liability. However, a risk-free discount rate is not called for, because even states with constitutional guarantees cannot be prevented from amending their constitutions and defaulting. For simplicity’s sake, we treat public pensions like private ones, while recognizing that the former’s unfunded liabilities are almost certainly greater than such treatment would indicate.
    13. “Actuarial liabilities” are the estimate of the present value of liabilities accrued by a pension plan. The calculation is based on actuarial assumptions about future actions by plan participants and employers—for example, growth in employee salaries (which affect the size of pension benefits), average retirement age, and average lifespan in retirement. These liabilities are then discounted to the present day through the use of a discount rate.
    14. We do not claim that this is a perfect estimation of asset values, as we do not have direct insight into the specific performance of plan assets since last reporting. While some plans invested a good portion of their non-stock assets in relatively safe bonds, many plans invested in hedge funds, real estate, or other investments that may have moved as much or more than the stock market. For funds that most recently reported asset values in 2009 (when markets were lower than today), we are likely overstating the funding gap; for those that last reported in 2008 or earlier, we likely understate it. However, we believe that our adjusted estimates are closer to today’s reality than the funds’ own reported actuarial asset values.
    15. Bureau of Labor Statistics, “National Compensation Survey: Employee Benefits in the United States, March 2009.”


    Bader, Lawrence N., and Jeremy Gold. 2003. “Reinventing Pension Actuarial Science.” The Pension Forum 14(2): 1–13.

    Biddle, RiShawn. 2010 (February 11). “Out of Chalk.” The American Spectator, available at

    Brown, Jeffrey R., and David W. Wilcox. 2009. “Discounting State and Local Pension Liabilities.” Paper submitted to AEA, session “Pensions and Health Care: Fiscal Challenges for State and Local Governments.”

    Costrell, Robert, and Michael Podgursky. 2010. “Golden Handcuffs.” Education Next Winter 2010 (10:1). Available at

    Diel, Stan. 2009 (November 12). “Alabama Board of Education Endorses Proposed Changes in K–12 Funding, Teacher Retirement.” Birmingham News.

    Fahey, Tom. 2009 (November 11). “Retiree Hike Hits with Thud.” Union Leader, available at

    Fitch, Stephane. 2009 (January 22). “Gilt-Edged Pensions.” Forbes, available at

    GAO (Government Accounting Office). 2008. “State and Local Government Retiree Benefits: Current Funded Status of Pension and Health Benefits.” Report GAO-08-223 to the Committee on Finance, U.S. Senate, available at

    GASB (Government Accounting Standards Board). 2009. “Invitation to Comment: Pension Accounting and Financial Reporting.”

    Girola, James A. 2007. “The Corporate Bond Yield Curve for the Pension Protection Act.” U.S. Department of the Treasury, available at

    Gittleson, Kim. 2010 (February 4). “Teacher Pension Fund Lost $9 Billion Last Year While Costs Rose.” Gotham Schools, available at

    Gold, Jeremy. 2000. “Assumed Rates of Discount for Valuations of Publicly Sponsored Defined Benefit Plans,” available at

    Jones, Norman L., Brian B. Murphy, and Paul Zorn. 2009. “Actuarial Methods and Public Pension Funding Objectives: An Empirical Examination.” Public Pension Finance Symposium (May 2009), session 2, “The Rational for Traditional Actuarial Models.”

    Jourgensen, Thor. 2009 (November 12). “McGee: Sans Reform, Pension Plan Ruinous to Lynn, Others.” The Daily Item, available at

    Monahan, Amy B. 2009. “Legal Limitations on Public Pension Plan Reform.” Paper at conference “Rethinking Teacher Retirement Benefit Systems,” Nashville, Tenn., February 19–20, 2009.

    NASBO (National Association of State Budget Officers). 2009 (December). Fiscal Survey of States, available at

    Nash, James. 2010 (January 3). “Taxpayers Asked to Cover Rising Pension Costs for Government Employees.” Columbus Dispatch, available at

    Novy-Marx, Robert, and Joshua D. Rauh. 2008. “The Intergenerational Transfer of Public Pension Promises.” NBER Working Paper 14343, available at

    Pew Center on the States. 2006. “Promises with a Price: Public Sector Retirement Benefits.” Available at

    PSERS (Pennsylvania Public School Employees’ Retirement System). 2009 (December 11). “An Update on PSERS Rate Spike/Plateau Presentation,” available at

    Rispoli, Michael. 2009 (March 16). “Pension Deferral Bill Passes N.J. Legislature.”, available at

    Waring, Barton. 2009. “A Pension Rosetta Stone: Reconciling Actuarial Science and Pension Accounting with Economic Values.” Public Pension Finance Symposium, Society of Actuaries, May 4, 2009. Available at

    Witte, Brian. 2009 (October 20). “O’Malley Asks Superintendents to Scour for Savings.” Baltimore Sun, available at .