June 1, 2020

Pension Plans And The 'Duty Of Impartiality' Rule: Implications For Plan Design

“Now more than ever, investment leaders have the opportunity to make life-changing differences for roughly four billion people’s savings and investments. They will do so by drawing from the widest range   of leadership skills to manoeuvre through the epic challenges this crisis presents and by emerging with stronger, fairer, and more sustainable organizations.”

Roger UrwinTOP1000FUNDSMay 14, 2020

“Our Funding Management Policy provides the sponsors with a guidance framework for decision-making when there is a funding surplus or shortfall. Funding zones provide a point of reference for whether  action is required…..answering questions of when it is prudent to increase or decrease benefits, raise or lower contribution rates, or simply conserve assets for an uncertain time. Ultimately, it is the sponsors’ responsibility to decide what actions to take…..”.

Ontario Teachers’ Pension PlanAnnual Report 2019

Financial Individualism has a Cost

In his recent article in The Economist, Mark Carney reminded us that financial individualism has a cost and that only collective action can deliver good healthcare, financial security, and a livable planet for all. In a new article in TOP1000FUNDS, Roger Urwin reminds us that it also requires effective organizations with strong leaders to actually deliver good healthcare, financial security, and a livable planet. This Letter expands on those themes, specifically on its financial security dimension. Past Letters have addressed the cost of financial individualism if people are left to create and execute a retirement savings plan on their own. Logic suggests and research confirms that while some people can do this successfully, most cannot. For the vast majority of people, the problem begins with failing to systematically save for retirement and invest those retirement savings skillfully and cost-effectively. It continues with failing to use the accumulated retirement savings skillfully and cost-effectively to finance life after work.

How can we best help this vast majority of underinformed and behaviorally-challenged people? Logically, by designing and building collective retirement income systems (RIS) that do most of the heavy lifting for them. The ideal RIS has three pillars: 1. A universal tax-funded base pillar, 2. A workplace-based pillar with a variety of large, well-managed workplace pension plans (WPPP) designed to serve different sectors of the workforce, and 3. A do-it-yourself pillar ideally only used by people with the requisite financial planning and investment skills.    

Empirical evidence confirms this design logic is sound. Countries that have a sustainable Pillar 1 and strong workforce participation in large, well-managed Pillar 2 WPPPs get top scores in the Melbourne-Mercer Global Pension Index (MMGPI). Countries with lower scores tend to have two kinds of problems: 1. Their Pillar 1 programs are unsustainable in the decades ahead, and 2. The participation rate in effective Pillar 2 WPPPs is low, which means by default, high participation rates in various ‘for profit’  Pillar 3 offerings or no retirement savings at all.  ‘For profit’ Pillar 3 offerings often reflect the asymmetric information reality of the customers paying too high fees for too little value. This can easily result in an ultimate 50% shortfall in retirement income relative to participating in an effective Pillar 2 WPPP.i

Designing Effective Pillar 2 WPPPs

So designing, building, and maximizing participation in effective Pillar 2 WPPPs is a critical RIS success driver. Starting with design, the first step is to ditch its dysfunctional ‘DB vs. DC’ framing. People’s financial lives have three phases: 1. The pre-work phase prepares people for the work phase, 2. In the work phase, a proportion of income earned is saved and invested to finance the post-work phase, and 3. In the post-work phase, the accumulated retirement savings are converted into an income for life stream. Ideally, the savings rate during the accumulation phase is made affordable by earning a high investment return on the accumulating retirement savings, and these savings generate a predictable lifetime pension during the decumulation phase by adopting a more conservative investment policy.

All this suggests three requirements for a country’s Pillar 2 RIS segment to be effective: 1. Maximize Pillar 2 participation by requiring employers to enroll their employees in a qualifying Pillar 2 pension arrangement with a realistic contribution rate, 2. Create pension plans to accommodate the accumulation and decumulation phases of members’ life-cycle journeys, and 3. Create large-scale pension management organizations that can competently manage these life-cycle journeys in the best interests of their participants.

The design logic set out above suggests separate accumulation and decumulation money pots, with member assets shifting from the accumulation pot to the decumulation pot as members approach, and move into the post-work phase. Is there a country that actually has a Pillar 2 RIS segment with separate accumulation and decumulation pots? Australia comes closest to having adopted the 2-pot model, and it achieved the 3rd-highest ranking out of 37 countries in the 2019 MMGPI.ii            

What about 1-Pot WPPPs?    

The 2-pot WPPP design solves the affordability/safety investment dilemma for retirement savers. The retirement savings of young workers all go into the accumulation pot, a default mechanism starts to transition retirement savings into the decumulation pot as workers age, and the decumulation pot (with longevity risk pooling) begins to pay out reasonably predictable lifetime pensions when workers retire.iii From a fiduciary duty perspective, such a design, effectively implemented, clearly passes the ‘impartiality test’ in the sense that it considers the financial interests (including differing risk preferences) of both workers and retirees.

However, most traditional WPPPs with decades-long histories in Europe and North America were set up in the 1950s and 1960s as 1-pot fully-guaranteed DB plans. While in the early years these plans had young memberships and positive cashflows, that is no longer the typical case. They now have mature memberships with significant proportions of older workers and retirees. At the same time, real interest rates are much lower. This combination of aging demographics and lower investment return prospects has triggered two types of responses: 1. Many corporate sponsors of traditional DB plans have closed them, and 2. Many public sector/organized labour plan sponsors have stayed with collective 1-pot WPPPs, but have dropped the historical benefit guarantees typically attached to DB plans. Modern versions still target a defined benefit, but no longer with a sponsor guarantee.

The fiduciary duty of impartiality raises four 1-pot Target Benefit (TB) Plan design questions: 1. Who are the risk bearers? 2. If the Plan goes ‘off course’, what tools can be used to bring it back ‘on course’? 3. How will those tools actually be used in an ‘off course’ scenario? And 4. Does the structure meet a reasonable fairness test that one group of Plan participants is not systematically favoured at the expense of another group?

The ‘who?’ question is straight-forward. At any point in time there are three parties involved: 1. the sponsor group, 2. the active workers group, and 3. the retirees group. Of course, with the passage of time, the membership of those groups will change (100% on a generational basis). The ‘tools?’ question is also straight-forward. There are two: 1. Benefits can be raised or lowered, and 2. Contribution rates can be raised or lowered. The ‘how?’ and ‘fair?’ questions are the difficult ones. Which tools will be used and how will their use impact (a) the three parties at a point in time, and (b) the following generations of these parties? Also, how is the affordability versus safety dilemma fairly resolved over time? These questions are best addressed through a case study.

How has OTPP been Addressing these Questions?

The Ontario Teachers’ Pension Plan (OTPP) manages a TB Plan for 329K members, of which 184K (56%) are active (average age 43) and 145K (44%) are retirees (average age 72). The plan sponsors are the Ontario Teachers’ Federation and the Government of Ontario, who are jointly responsible for setting plan benefits, contribution rates, and addressing plan surpluses and shortfalls. A key ‘pension deal’ element is that the established contribution rate is to be split 50-50 between active teachers and Ontario taxpayers.

The OTPP Board of Directors is appointed by the plan sponsors, but describes itself as professional, independent, and required to act in the best interest of all beneficiaries. To that end, it played a major role in persuading the plan sponsors to introduce conditional inflation protection (CIP) into the plan benefit formula over the course of the Global Financial Crisis (GFC). Why? Because the ‘duty of impartiality’ principle requires that when the Plan asset value is materially below the Plan funding target, righting the ship should not fall wholly on contribution increases by active members and taxpayers. Retirees should bear some of the burden as well by foregoing a portion of inflation-linked pension increases until full funding is reestablished.

The OTPP Board also plays a more direct role in ensuring Plan fairness. One of its key responsibilities is to set the discount rate to be used to calculate Plan’s funding target. How does a Board set a funding discount rate ‘in the best interest of all beneficiaries’? A ‘too high’ rate would favour current plan members and taxpayers at the expense of future ones, and vice versa, a ‘too low’ rate would do the reverse. A ‘fair’ rate lies somewhere in between..... starting with the term structure of default risk-free interest rates, the maturity of plan membership, a cushion for major adverse events, and adding ‘reasonable’ increments reflecting a premium for risk-taking as well as for OTPP’s expected skill in generating excess returns net of expenses.

What has the outcome been of this deliberation process at OTPP over the course of the last 10 years? Table 1 tells the tale: the OTPP Real Funding Discount Rate has declined steadily in line with falling interest rates and the rising maturity of Plan membership. Despite this declining discount rate, note that the Plan’s funded ratio (i.e., the ratio of smoothed Plan assets to the Funding Target calculated with the Funding Discount Rate) did not dip below 100%. However, the 2009, 2011, 2014, and 2016 Funding Target calculations included reductions due to invoking the CIP option. Full inflation indexation was restored in the calculation of the 2019 Funding Target calculation.     

Table 1 OTPP  Real Funding Discount Rates and Funded Ratios

  2009 2011 2014 2016 2019
Real Funding Discount Rate 3.65% 3.25%  2.85%   2.75% 2.60% 
Funded Ratio (including the CIP impact) 101%   100%  101% 102%  103% 

Source: OTPP Annual Report 2019

Looking ahead, through its Funding Management Policy (FMP), OTPP has a clear roadmap on how it should respond to future Funded Ratio changes. There are five zones: 1. Usable Surplus, 2. Temporary Plan Improvements, 3. Funded Zone with Base Provisions, 4. Funded Zone Without Base Provisions, and 5. Shortfall Zone. The Plan went into Zone 4 after the GFC (leading to only partial inflation indexation) and has recently recovered back into Zone 3 (fully funded with base provisions). The benefit of having a FMP is clear: no ‘after the fact’ negotiations about how funding surpluses or deficits will be allocated. Finally, what about OTPP’s investment policy over the course of the last 10 years….is there evidence that it has been reducing balance sheet risk as its plan membership continues to slowly age? The 7% shift out of  equities and the 9% shift into bonds over this period suggests a ‘yes’ answer.

Lessons Learned

The broad lesson from the OTPP case study is that it is indeed possible to manage 1-pot TB plans with a ‘best interests’ mindset, but it is not easy. Key success elements include:

  • A plan design that shares funding shortfalls and surpluses fairly and transparently between plan participants.
  • A clear split of the roles and responsibilities of the plan sponsor(s), the plan board of directors/trustees, and plan management.
  • The role of the board is especially critical in enforcing the ‘best interests/impartiality’ rule. It means ensuring the plan’s funding target is based on assumptions and risk exposures that are ‘reasonable’. That means (a) they reflect current economic and plan maturity realities, and (b) they have a conservative bias.
  • These board responsibilities have strong implications for board composition. In addition to understanding the implications of the ‘best interests/impartiality’ rule, it requires people with strategic thinking capabilities, and a collective skill/experience set that includes investment, actuarial, HR, IT, and communications/reporting.
  • A regulatory regime that is principles-based rather than rules-based.iv

In conclusion, the time has come to bury to tiresome ‘DB vs. DC’ design debate. The 2-pot and 1-pot WPPP models set out above can both be made to work in the best interests of all plan stakeholders. But only if sensible legislation, regulation, good governance, and effective management are added to the mix. And to repeat Roger Urwin’s words on page 1: only if the leaders of the global pension and investment communities recognize that now is the time to step up and play constructive roles in making this ‘in the best interests’ transition happen. 

Keith Ambachtsheer  


  1. See the October 2018, July 2019, and January 2020 Letters for more on these realities.
  2. Australian employers must contribute 9.5% of pay on their workers’ behalf into a qualifying super fund. At about AU$3T for 15M workers, the average member account balance is about AU$200K. A recent review is leading to a number of measures to improve the super system.
  3. Ideally, it would also be possible to integrate the longevity risk pooling feature of a country’s Pillar 1 RIS component, with the flexibility of keeping individual ownership of Pillar 2 retirement savings. See, for example, Munnell and Pozen, “401(k) Nest Eggs Can Help Get the Most from Social Security”, P&I, Feb 24, 2020.
  4. For example, the complexity and unintended consequences of the current rigid rules-based pension regulation system in the Netherlands has been causing confusion, dissatisfaction, and distrust there for some time. At the other extreme, the USA has no regulation covering public sector pension plans at all. That, combined with political influence on many boards of trustees, has often led to disregard for the ‘best interests/impartiality’ rule. An example is the use of Real Funding Discount Rates materially higher than the OTPP rates set out in Table 1.


KPA Advisory Services is pleased to share this edition of The Ambachtsheer Letter with all readers; if you wish to become a KPA Advisory Client/gain access to ALL Letters, please see the Services page on our website.

The information herein has been obtained from sources which we believe to be reliable, but do not guarantee its accuracy or completeness. 

Advisory Service clients have access to full issues of the Ambachtsheer Letter.

Become an Advisory Service Client
or Login
Back to Top