Public Pensions and the Perfect Storm: National Strategies for Reform

By Barbara Reid

By Barbara Reid

All across the country, states are grappling with finding solutions to sustain their retirement systems. In February 2010, the Pew Center on the States issued a report titled The Trillion Dollar Gap: Underfunded State Retirement Systems and the Roads to Reform, analyzing the nationwide gap between the $3.3 trillion in public pension benefit obligations and the $2.3 trillion set aside to pay for those obligations. The report examined the roots of the underfunding problem and identified the pressures that created, and continue to contribute to, the funding gap. It also identified the reform measures that are gaining momentum in many states to address those unfunded liabilities. Finally, each of the 50 states received a grade on management of their public sector retirement benefit obligations, with grades of “solid performer,” “needs improvement” or “serious concerns.”

The initial release of the Pew report attracted immediate media attention, particularly in New Hampshire where the management of the long-term pension liabilities fell into the “serious concerns” category. Since then, however, little has been mentioned about that grade or the options that would help to improve the financial health and stability of the system. Before reviewing some of the reform measures highlighted in the report, let’s first look at the current financial status of the New Hampshire Retirement System (NHRS) and recap some of the major contributing factors that led to the present condition of the system.

Funding Ratio

As stated in the NHRS June 30, 2009 summary annual report, “the primary measure of a benefit plan’s progress is its funded status that compares the assets available to the benefits that must be paid.” Chart A illustrates the 10-year funded ratio history of the NHRS, indicating a high of 89.9 percent in 2000 and a low of 58.3 percent in 2009. Note, however, that the funded ratios for the fiscal years prior to 2007 were calculated using a different cost method and, therefore, are not comparable on a consistent basis. What exactly does the current 58.3 percent funding ratio mean? It means that the assets available are only sufficient to cover 58.3 percent of the liabilities. In most financing arrangements, the ultimate goal is to have assets sufficient to cover 100 percent of the liabilities. Yet, for a pension system, a funding ratio of 100 percent is not necessarily a realistic benckmark because all benefits are not due and payable at one time, and generally stretch over a 25-to 30-year horizon. Therefore, the more common benchmark for measuring the health of a pension system is generally 80 percent. According to the Pew report, in the aggregate, state systems were 84 percent funded in 2008, with 21 states falling below that benchmark of 80 percent. New Hampshire, with a 67.8 percent ratio in 2008 and a 58.7 percent ratio in 2009, fell well below the benchmark.

A Perfect Storm

The old adage “the squeaky wheel gets the oil” could not be more true in terms of identifying the major contributing factors that have led to the current underfunded status of the NHRS. The years when a wealth of investment earnings appeared to be flowing into the system made for easy decisions to expand benefits in the forms of cost-of-living adjustments, medical subsidies and statutory changes that provided the opportunity for employees to maximize their pension annuities. But, several years ago, initially as a result of changes in pension accounting standards, the wheels of the system began to squeak rather loudly, forcing more focused attention on the policies, procedures and operations of the system by the primary stakeholders—those for whom the system was designed to provide benefits, and those for whom the system was designed to pay for those benefits. As a result of this closer scrutiny, the convergence over many years of a perfect storm involving the actuarial methodology, gain sharing and investment losses finally came into view.

Actuarial Valuation

Acknowledging that there were a variety of methods for valuing the assets and liabilities of public pension systems, the Government Accounting Standards Board (GASB) decided that, for financial reporting purposes, consistent reporting among the hundreds of public pension plans was necessary in order to provide comparable data on which to base public policy decisions regarding these systems. Through Statement 50, in 2007, the GASB promulgated the use of the Entry Age Normal (EAN) valuation method, the method then used by the majority of public pension systems but which required a handful of systems, including New Hampshire, to change. When this new valuation method was implemented in New Hampshire, the primary flaw of the previous valuation method, known as the Open Group Aggregate (OGA) method, came to light.

The OGA method was implemented in the early 1990s in an attempt to help ease state budgetary pressures and appears to have been intended to be a short-term solution for holding down pension costs. Unfortunately, the OGA methodology remained in place for 15 years, with few stakeholders recognizing the primary flaw in this methodology: it masked the true financial condition of the system, resulting in artificially low employer rates over the years it was in effect. Once this mask was removed in 2006, an unfunded actuarial accrued liability of $2.7 billion was revealed. Essentially, due to the valuation methodology used, employers had been undercharged—and underpaid the system for years.

Gain Sharing

During the time when the true financial picture of the system was blurred by the OGA methodology, a “gain- sharing” provision in the law further influenced the funding ratio. Through the gain-sharing provision, when investment earnings were greater than .5 percent of the assumed rate of investment return, these so-called “excess” funds were indirectly diverted to a special account, which was then used to pay for additional benefits such as cost-of-living increases and medical subsidies for retirees. Although not an entirely inappropriate funding mechanism for such additional benefits, gain-sharing provisions are only a reasonable funding option when (1) the public pension system is very well funded, with a funding ratio near 100 percent, and (2) the trigger rate for computing the amount of excess funds is sufficiently higher than the assumed rate of return, in the vicinity of at least 1.5 percent or 2 percent higher. However, the New Hampshire pension system did not meet either of these criteria—it was not well funded, and the trigger was only .5 percent above the assumed rate of return.

The main problem with the gain-sharing provision was that it diverted investment earnings during the good years when earnings were high, earnings that should have been held in the corpus of the pension fund to offset the bad years when actual investment earnings were lower than expected, or when investment losses occurred.

Chart B, again from the NHRS June 30, 2009, summary annual financial report, shows a 10-year history of the time-weighted annual returns. From 2000 to 2007, the assumed rate of return was 9 percent. As shown, in several years (2000, 2004, 2005, 2006, and 2007) the earnings exceeded 9 percent. However, there was one year in which investment earnings were below the assumed rate (2003) and several years which experienced investment losses (2001 and 2002). If investment earnings above 9.5 percent are diverted, then an assumed rate of 9 percent (an assumption of the average rate which will be achieved over a period of time involving both good years and bad) can never be achieved unless the actual rates of return never fall below 8.5 percent.

In other words, with the gain-sharing provision, the assumed rate of return of 9 percent was nearly impossible to achieve. Yet, for years, the 9 percent rate was used to estimate the amount of investment earnings that would be available to pay for pension obligations. So, while an assumed rate of 9 percent may have been on the high side, at the time, it was not necessarily an unreasonable assumption of the average return over time. The fact that funds were diverted through the gain- sharing provision caused the 9 percent assumed rate to be unreasonable and unattainable. And again, along with the mask of the OGA methodology, an understanding of the long-term impact of gain sharing appears to have been missed by many stakeholders until fairly recently.

Investment Losses

There are only three sources of funding for the pension system: employee contributions, employer contributions and investment earnings. While employee contribution rates are set in statute (generally 5 percent for Group I, and 9.3 percent for Group II), employer rates fluxuate every two years based upon an actuarial valuation of the assets and liabilities. Investment earnings generally provide 60 percent to 70 percent of the funding for a public pension system, while the employer and employee contributions combined provide 30 percent to 40 percent.

When actual investment earnings miss the mark (that is, when investment performance is lower than the assumed rate of return), the difference is made up by increases in the employer rates. Regarding the investment losses of 2008 and 2009 (-4.6 percent and -18.1 percent, respectively), it’s important to note that these losses are not yet reflected in the rates employers are currently paying due to the timing of both the biennial actuarial valuation and the rate setting process. These losses will be reflected in the FY 2012 and FY 2013 rates. The estimated rates for this two-year period, which begins July 1, 2011, were presented to the NHRS Board of Trustees at their November 10, 2009 meeting. (These estimated rates and historical employer rates since the inception of the system are available in table and graph form.) Note that these rates reflect the total employer rates (combined state and local share) for teachers, police and fire, rather than just the local share (70 percent for 2010, 75 percent for 2011 and 65 percent for all other years).

Also, it’s important to know that NHRS uses a smoothing process in which investment gains and losses are recognized over a number of years. Therefore, the impact of the 2008-2009 investment losses will be spread over a five-year period, so as investment returns improve, those gains will be offset over the next several years by a portion of the recent losses.

Reforms Highlighted in the Pew Report

According to the Pew report, effective reform measures as reported by 231 public pension systems across the country generally fell into five categories and offer the opportunity to improve the performance of public sector retirement systems in both large and small ways.

1. Keeping up with funding requirements. This entails honoring funding requirements in both good times and bad. It also requires the use of accurate assumptions, one of the more significant being the assumed rate of return. In 2008, the assumed rates of return ranged from 7.25 percent to 8.5 percent.

2. Reducing benefits or increasing retirement age. Reforms included increases in retirement age by as little as two years, increasing the years of service and capping the amount of overtime used in calculating benefits. These changes were most often enacted for new hires, although several states implemented such changes to current employees. Several states require that any proposal that affects pension benefits or costs receive a “full actuarial analysis to determine the long-term price tag.”

3. Sharing risk with employees. This entails designing pension plans that share the risk of investment volatility with employees. This is accomplished by enacting benefit systems that combine elements of both defined benefit and defined contribution plans, resulting in what is now referred to as “hybrid” plans.

4. Increasing employee contributions. Several states have enacted recent increases to employee contributions. One state requires that employers and employees pay equal shares of the required contributions, which “tends to increase the attention that employees give to the health of the pension system and increases pressure to keep it well funded.”

5. Governance and investment oversight. Several states have sought to “professionalize the complex task of pension investments by shifting oversight away from boards of trustees to specialized bodies that focus on investments.” The report also noted that several pension reform commissions are considering changes similar to Oregon’s 2003 reforms, “heightening qualifications for trustees and shifting membership so that boards are not dominated by pension beneficiaries.”

System Changes

Over the past four years, there have been a number of significant changes to the NHRS addressing the conditions that led to the perfect storm, with some of these changes highlighted in the Pew report. As previously mentioned, the actuarial valuation method was changed to EAN, shining a brighter light on the financial status of the system and its true funding requirements and, as a result, employers began paying, through their contribution rate, the $2.7 billion unfunded liability over an amortization period of 30 years. The assumed rate of return was lowered from 9 percent to 8.5 percent. The gain-sharing provision was amended so that the diversion of investment earnings to the special account will occur only when earnings exceed 10.5 percent, provided that the funding ratio is at least 85 percent, a level which, based on the current ratio of 58.3 percent may not be achieved for at least 20 years. Changes were also made regarding the qualifications of the NHRS board members, now requiring expertise in the field of business, finance or institutional investment, along with the creation of an independent investment committee.

Yet, despite these recent changes, NHRS pension liabilities continue to increase at an unsustainable rate. From 2008 to 2009 alone, pension liabilities increased more than 8 percent, from $7.8 billion to $8.5 billion. An 8 percent increase each year eventually results in a doubling of the liabilities in just 10 years, and an 8 percent annual increase is not unrealistic considering, at a minimum, the expected impact of baby boomers on the pension system in the coming years.

While 45 pieces of legislation dealing with the NHRS were enacted over the past four years, only one (repealing the option to purchase non-qualified service credit) had a direct impact on the growing liabilities. Much of the legislation dealt with administrative, operational and governance changes, or provided additional or expanded benefits. Additional reform measures such as those enacted by other states will be necessary to address the growing liabilities of the NHRS, to help ensure the long-term health and solvency of the system for past, present and future employees, and to make this system affordable for local government employers and, ultimately, New Hampshire taxpayers.

Barbara Reid is government finance advisor for the New Hampshire Local Government Center. She can be reached at 800.852.3358, ext. 145, or by e-mail at


Pew Center on the States, The Trillion Dollar Gap: Underfunded State Retirement Systems and the Roads to Reform

New Hampshire Retirement System, 2009 Summary Financial Report

Additional charts and resources can be found on LOGC's website