Lessons From Fully Funded Defined Benefit Pension Plans

With another round of public pension fixes imminent, what are the keys to having healthy traditional defined benefit plans in this day and age?  Some answers to that question can come from an unexpected place:  the retirement plans Minnesota public sector unions offer their own employees.

“Urgency” is a word that doesn’t get used much in the world of Minnesota public pension policy.  The primary reason is that major problems can fester for years before affecting government budgets.  It doesn’t help that stakeholders and their consulting experts have classically conditioned lawmakers with phrases like “on track for full funding in 30 years” and “long-term perspective” anytime they get even the least bit fidgety about pension health.

But even in the protracted timeframes of pension policy, things can move relatively fast.  For example, a report this summer from the Pew Charitable Trusts on government defined benefit pension plans identified Oregon as one of only three states with fully funded plans.  At the same time Pew released this report – based on 2014 data – Oregon’s public pension plans were reporting a unfunded liability for 2016 of $21.8 billion.  That’s the fallout from underperforming investment returns over the last two years combined with an acceptance that return expectations needed to be reduced.  As a result, schools, cities, state agencies and other public employers across Oregon will have to pony up an extra $885 million over the next biennium to fund public pensions – about 10% more than previously forecast and a 44% increase from the current $2 billion per biennium cost.

Given the veto of this year’s omnibus pensions bill, Minnesota lawmakers will be addressing our own unfavorable investment results1 and the prospect of adjusting economic assumptions when they revisit the issue of sustainability in 2017.  And it seems a sense of urgency on these matters is creeping into the legislature.  Senator Sandy Pappas, who will likely have the pension commission gavel next session if the DFL retains control of the Minnesota Senate, has said she will “put forward a long term fix.”2  The unresolved issue is the nature of the “fix” – the reforms lawmakers will enact to reestablish a healthy traditional defined benefit pension plan.

Hints at what such measures might be necessary can be gained by examining the practices of defined benefit pension plans that have maintained full funding through the recent economic and investment turmoil.  Given the bleak condition of public pension plans around the country, it seems appropriate that such an investigation is best based on whatever points of light exist in the dwindling universe of private sector defined benefit plans.  Our review of annual reports filed with the IRS finds some plans very close to home worth considering.  Two such plans are the Education Minnesota Employees Pension Plan and the AFSCME Employees Pension Plan.

“One of these things is not like the other… ”

As a private, single-employer defined benefit plan, the Education Minnesota Employees Pension Plan files actuarial information annually with the Internal Revenue Service.3  According to its latest filing the plan serves 144 active employees, 160 retirees and their beneficiaries, and 33 vested participants no longer employed by the union.  Unlike the struggling TRA plan in which the union’s own membership participates, the pension plan providing retirement security for Education Minnesota employees features a funded ratio of 133.6%, which represents a surplus (current assets in excess of the present value of future liabilities) of approximately $20 million.

The considerably healthier condition of the Education Minnesota pension plan is not a function of being less generous plan than its TRA equivalent.  In fact, it’s just the opposite.  The Education Minnesota employees plan substantially exceeds the TRA plan covering the union’s members on both eligibility features and generosity dimensions – and 100% of contributions come from Education Minnesota via member dues, with no contributions from employees themselves.  The one big design difference is that, unlike TRA, Education Minnesota does not seem to offer any post-retirement cost of living adjustments.  However, since Education Minnesota employees contribute nothing from their own paychecks to support their retirement plan, they are more able than TRA members to establish their own independent retirement accounts to protect against the effects of inflation during their retirement years.

Comparison of Selected Features, Teachers Retirement Association and Education Minnesota Employees Pension Plans

Teachers Retirement Association Education Minnesota Employees Pension Plan
Funded Ratio* 79.97% 133.6%
Final Average Compensation Based on highest 5 earning years Based on highest 3 earning years
Monthly Accrued Benefit 1.9% of final average compensation 2.0% of final average compensation
Normal Retirement Age 67 62
Early Retirement “Rule of 90” for members enrolled before July 1, 1989; none for more recent participants Attainment of age 55 with 5 years of vesting service
Total Annual Contribution as a percent of salary 15.97% Estimated at approximately 20%**
Employee share of contribution as a percent of salary 7.5% Zero – employees do not contribute
Special Retirement Benefit None Available to executive staff employees, associate executive staff employees and officers
Discount rate used to calculate present value of  plan liabilities in determining funding requirements 8.44% 6.45%
* TRA funded ratio on market value basis, as of 7/1/2015.  Market value basis chosen for comparative purposes as Education MN employee plan reports asset values on a market value basis.  Education Minnesota funded ratio based on valuations as of 9/1/2014.  Independent auditor's report for 9/1/2015 shows funded ratio of 129.6%.
** Estimate based on matching total contributions reported on Form 5500 Schedule SB with salary totals reported on Education Minnesota’s tax return (Form 990).

Sources:  Education Minnesota Employee Plan Form 5500 Schedule SB (March 2016) and accompanying independent auditors report (Harrington Langer & Associates); 2015 TRA Actuarial Valuation Report

How is this combination of solid financial health and comparatively more generous benefits possible?  The answer doesn’t lie in the investment portfolio.  Notably, the Education Minnesota Employee Plan asset schedule is completely devoid of the whiz bang (and relatively expensive) “alternative investment” products which have become an increasing staple of public pension funds across the nation in their scramble for higher yield.  Rather, the portfolio reflects what might be found in your highly cost-conscious grandmother’s Schwab statement – a collection of very low cost mutual funds.  Nearly 40% of their assets are in simple index funds.

The answer appears to be rather simple – they pay for it.  We estimate that Education Minnesota’s contributions to its employee pension plan are about 20% of salary.  Compare that to the 15.97% contribution rate for TRA – which comes from school districts and employees.  Four percentage points may not seem like much of a difference, but for TRA that amounts to $189 million in additional investable assets in just this past year.  Considering that TRA has experienced a contribution deficiency (where contributions have fallen short of what their actuaries say are necessary to pay for new benefits and to amortize unfunded liabilities) in every year since FY 2006, it’s pretty clear to see how having one-third higher contributions over the last decade would have made their current valuation reports look a little different.

The pension plan for the American Federation of State, County, and Municipal Employees (AFSCME) is classified as a multiple employer defined benefit plan – serving the employees of individual councils across the nation but funded as if a single employer employed all the participants.  Like its educational sibling, this union employee pension plan is also in excellent shape with a funded ratio of 116.87% as of its most recent available Schedule SB filing dated July 23, 2015.

Because it is a multi employer plan and its Minnesota members represent union employees in both the state employee (MSRS) and local employee (PERA) plans, a side by side comparison is beyond the scope of this article.  Like the Education Minnesota Employees Plan, the AFSCME plan is also relatively generous compared to the plan in which those they represent participate (e.g.  final average compensation based on “high 3” salary with an accrual rate of 2.4% of final average compensation).  However, there are some interesting differences between the two plans:

Notable Differences Between AFSCME and Education Minnesota Employees Pension Plans

AFSCME Education Minnesota Employees Pension Plan
Employee contributions 6% None
Normal retirement age 65 62
Reduced benefits for early retirement? Yes No

In addition the AFSCME plan has some interesting features that may also assist in ensuring its ongoing sustainability:

  • A cap on pensionable earnings (it’s a huge cap, but still a cap nevertheless)
  • A post retirement cost of living adjustment equal to only one-half of the Consumer Price Index 
  • A benefit rate for future years of service that seemingly adjusts to reflect fund conditions and can actually decline if necessary

One key element is similar: the discount rate AFSCME uses to calculate the present value of the plan’s current liability to determine funding requirements is also substantially lower (6.52%) than its public sector counterparts.

The ERISA Difference

So what explains the discrepancy in health between these private and public sector defined benefit plans?  The answer is largely attributable to old fashioned federal regulation – from which public plans are exempt but private plans are not.

Congress enacted the Employee Retirement Income Security Act of 1974 (or ERISA) to address the danger that terminating defined benefit pension plans with large unfunded liabilities would threaten millions of Americans’ retirement income.  A cornerstone of the legislation was the creation of the Pension Benefit Guarantee Corporation (PBGC) to backstop pension plan defaults.  But by 2005 dark clouds loomed on the horizon.  Faced with several large pension plan defaults and a PBGC deficit of $22.7 billion, an engineered bailout of the agency was becoming increasingly likely.

As a result – with a majority of 68% in the House and 95% in the Senate – Congress passed the Pension Protection Act (PPA) in 2006.  It is regarded as the most comprehensive reform of the nation’s pension laws since ERISA in 1974.  Lawmakers designed the PPA to increase the minimum funding requirements for defined benefit plans and strengthen the pension insurance system by correcting for several weaknesses of the existing rules.  According to a Congressional Research Service report4 those weaknesses included the following:

  • Sponsors of underfunded plans were not required to make additional contributions as long as their plans were at least 90% funded
  • Neither plan assets or liabilities were being measured accurately
  • Plans that were underfunded were sometimes able to amortize their funding shortfalls over periods as long as 30 years

Those weaknesses reflect policies that should seem rather familiar.

The PPA introduced new funding requirements for private sector defined benefit plans.  It established new rules for calculating plan assets and liabilities.  It obligated pension plans to calculate the present value of their liabilities based on bond rates that reflected the length of time until when those liabilities would need to be paid.  It required plans to be fully funded and required that any unfunded liability be amortized over 7 years.  And it established a new “at risk” category for defined benefit plans whose sponsors are required to make larger contributions to offset the greater liability of these plans.  The federal “at risk” threshold is 80% funded – which paradoxically is the same funding threshold many advocates have used to label public sector pension plans as “healthy.”

The Price Tag of Fiscal Responsibility

Ten years later, private sector sponsors that could deal with the new stringent demands of fiscal responsibility in providing these types of plans – like Education Minnesota and AFSCME – have kept them.  Those that couldn’t have either closed or eliminated them.  Trends suggest most private sector sponsors fall into that latter camp.

As this trend continues, the disappearance of defined benefit plans in the private sector has created a perception problem for public sector retirement advocates.  In their efforts to explain this discrepancy and justify the continued viability of DB plans, these advocates appear eager to throw the regulations designed to protect retirees and the public interest under the bus.  The National Institute for Retirement Security, a retirement research organization governed by public plan officials and state investment councils, blames the decline of the private sector DB plan on excessive regulation:

Private sector employers have been closing and freezing their pensions due to onerous laws and regulations enacted since the 1970s, including the Pension Protection Act of 2006. These rules created complicated funding rules, and increased contribution volatility when employers need steady, easy-to-estimate costs from year to year.

Who Killed the Private Sector DB Plan?  National Institute for Retirement Security, March 2011

This claim deserves two responses.  First, “spectacular” doesn’t begin to describe the irony of these advocates’ complaints about onerous, over the top regulation on this issue.  To criticize the federal government’s attempt to rid the private sector of the very practices that are now plaguing public sector DB plans across the country and placing state and local government budgets in such long-term fiscal jeopardy is a truly remarkable demonstration of a lack of self-awareness and the definition of chutzpah.

Second, the whole premise of having stable, predictable contributions while at the same time basing pension financing on riskier asset classes is deeply flawed.  If stakeholders insist on embracing higher risk investment strategies, accepting much greater contribution volatility has to be part of the deal.  Why?  Because higher risk investment strategies yield results that are inherently more volatile than a portfolio oriented toward lower risk fixed income investments.  This in turn makes the yearly changes in the required contributions also more volatile.  You cannot reach for higher expected returns to lower present funding costs and at the same time expect (or demand) “steady, easy to estimate costs from year to year."5

These ideas are mutually exclusive and portend trouble if you try to marry them (as we in Minnesota – along with others  – do).  As former State Board of Investment executive director Howard Bicker said, “If you look at our annual returns, very seldom have we been anywhere around 8.  It’s been 15 or 2 or 20. And that’s just the reality."6  Responsible contribution policy has to reflect that reality.  It’s also why Education Minnesota’s pension “surplus” (assets in excess of liabilities) reflects responsible defined benefit practice based on higher risk asset portfolios.  Having a surplus provides the funding cushion a pension plan with a more volatile investment portfolio needs to navigate difficult economic times.

So – what should Minnesota lawmakers do in 2017?  One course of action would be to adopt requirements for the state’s public sector pension plans that mimic the rationale and strategy behind the PPA – embrace more conservative investment and discounting assumptions, reduce the amortization periods for unfunded liabilities, and set contributions equal to what the system requires on an immediate (rather than phased in) basis.  Depending on how affordable these changes are to lawmakers, structural tweaks like those found in the union plans could be necessary – caps on pensionable earnings, reductions in benefit accrual rates going forward, even further reductions or the complete elimination altogether of post-retirement cost of living increases.  It would also mean resisting benefit increases and contribution cuts after these measures restore the plans to full funding to allow a reserve to be created.

To say this policy agenda would “trouble” most every pension stakeholder (including taxpayers) is a world class understatement.  The direct costs would likely require new taxes, redirection of tax dollars away from existing programs, or both.  As a result, the best odds are that lawmakers will once again tweak the system in a necessary, but ultimately insufficient, way.

Minnesota’s defined benefit pension plans are without question a high quality retirement benefit for public employees.  They are also without question a lot more expensive that we currently choose to admit – a reality we have to come to grips with.  The real urgency surrounding our pursuit of a lasting, fiscally responsible pension fix in 2017 revolves around an understanding and acceptance of that latter truth.

  • 1 According to the State Board of Investment, the “combined funds” in which pension system assets are invested had a return of -0.1% in FY 2016.
  • 2 "Veto Still Bugs House Republicans", Capitol Report, October 3, 2016.
  • 3 Sections 104 and 4065 of the Employment Retirement Income Security Act (ERISA) of 1974 require filing of Form 5500, Schedule SB.  Unless otherwise noted, all descriptive information for the Education Minnesota Employees Pension Plan comes from their latest federal filing dated March 18, 2016.
  • 4 Summary of the Pension Protection Act of 2006, CRS Report for Congress, October 23, 2006.
  • 5 Many argue that our current practice of “actuarial smoothing” in which investment gains or losses are phased in over five years is the mechanism by which pension plans can obtain the benefits of both lower cost plan funding (via investments in higher risk assets) and more stable, predictable contribution policies.   Smoothing could serve this purpose if 1) public plans were required to make the full actuarially required contribution each year, and 2) the liability discount rate was appropriately conservative.   Since neither of those preconditions exist in Minnesota, actuarial smoothing practices amount to nothing more than hiding the true funded status of plans while protecting weak governance structures by making it more difficult to enhance benefits or cut contributions in the wake of good investment years.
  • 6 “And Many Happy Returns: Howard Bicker Retiring” Teachers Retirement Information Bulletin, Fall 2013.