Think Tank Session #5 | Decumulation Institute

The Decumulation Institute organized its fifth semi-annual Think Tank Session in April of 2016. The participants included actuaries Don Ezra, Malcolm Hamilton; Bill Chinery; Clive Morgan, Paul Purcell, and fund governance leader Tom Iannucci and Dr. John T. Por as session organizer and facilitator.

Of the participants, three are still actively involved in the running and/or advising large public sector pension plans, and two (Chinery and Por) are currently serving as trustees for such plans.

Prior to the session, Decumulation Institute prepared a set of pension finance propositions for discussion.

These propositions represented a view of DB pension finance, currently followed, to the best of our knowledge, by none of North American publics sector pension plans or retirement systems. The propositions were prepared to start discussions without any expectations of general agreement on all points.

The propositions outlined for analysis during the Think Tank Session were as follows:

The value of future payments (pension plans, foundations, endowments, etc.) is bond-like and, equally important, often guaranteed (even if tacitly), not unlike deferred annuities issued by financial institutions.
Financial institutions use the expected returns of a “minimum-risk portfolio” – a bond rate of sorts – in pricing annuities. They must do so to ensure that they will not fail as institutions in case of adverse investment experience of the portfolio undergirding the promise.
There is no fundamental difference in the finance considerations between, say, an insurance company and that of a pension plan (either corporate or private) with the exception of public pension plans where though taxation fiat governments can force third parties – current and future taxpayers – to underwrite the promise.
Using the expected rate of return of an equity-dominated portfolio: (a) under prices the value of the promise, thus distorting trade-offs, leading to excessive investment risk taking (desired returns — and not risk appetite or tolerance — drive the portfolio’s construction), (b) for public pension plans makes intergenerational equity nigh impossible, and (c) makes stable contribution rates nigh impossible, as well.
The only reason such a discount rate is used is to lower ongoing funding requirements, as equity investing by definition jeopardizes the security of future payments.
The “guarantor” of the public sector pension benefits – the taxpayer – is not represented in the bargains struck, as the one negotiating on their behalf has significant conflicts of interest.
It appears that the viability of the system is tacitly based that the taxpayer eventually saves the system if in serious trouble due to underfunding and less than expected rate of returns. This fact has not been communicated to the taxpayer or anybody else, for that matter.
Using a discount rate significantly higher than those of a “minimum-risk portfolio” leads to extensive risk taking and “books” the results prior to having them accomplished. It also guarantees to the beneficiaries a much higher return than is available in financial markets to everybody else
Any funding or asset/liability study using a discount rate as the expected rate of return of an equity-dominated portfolio will likely result in a probability of meeting the liabilities in the next 10-15 years of 40-50% with approx. 50 % probability of doubling of the contribution rate, as well.

Questions highlighted for discussion of the propositions included:

What does the Advisory Group suggest the proper discount rate should be, and why?
By what measure could a pension board arrive at an appropriate risk appetite?
What is the role and responsibilities of intellectually honest and conscientious pension trustees? Given their limited time, what are the key issues and risks a pension board should be reported on and discuss?

At the outset, participants acknowledged that current funding and investment practices are not likely to change in the near future.

Therefore, session objectives included discussing and establishing a well articulated professional view different from those implied by the propositions, how to set appropriate discount rates, how to establish risk appetite or capacity, and finally how pension trustees should fulfill their fiduciary role under current practices. The knowledge outcomes of the session are listed below.

On the Propositions and Current Practices:

There was agreement that:
- For pricing the benefits, the expected rate of return of a low risk portfolio is appropriate, similar to that of corporate DB plans, and
- Keeping funding costs (i.e. contribution rates for funding) acceptable, equities should dominate investment portfolios. Thus for funding purposes the current practice (i.e. the use of the expected rate of return of an equity dominated portfolio) is appropriate provided that such assumed rate is reasonable.
As corollary, as equity investing leads to significant risks in terms of benefit security and contribution rates levels and stability, eventually the consequences of such risk taking (magnitude, bearer, and corrective actions) have to be clarified and communicated to plan sponsors and other affected parties.
Absent of such clarity, (i.e. spelling out who bears the risks and who receives the rewards for such risk taking), uncertainty exists about dealing with eventual negative consequences and also understanding of the affordable level of risks.
Current language around risk discussions treats such risks as if they were taken by abstract institutions, and not specific parties such as beneficiaries (active and retired) and tax payers.
Depending on risk and cost sharing arrangement, one party (tax payers) bears considerable risks without being aware of the level and the probability of the risks they bear and, equally important, not fairly rewarded for such risk taking (unlike the treatment of shareholders in private sector DB plans).
Some of participants pointed out that benefit guarantees have already by and large disappeared. Others – the majority– were of the opinion that it is not so as the deliberations and, more importantly, communications to members frequently use language and terms such as safe and secure, which absent feedback would in the beneficiaries’ mind probably add up to the same thing as guarantees. Such terms, would not, it was further argued, lead to beneficiaries’ believing that their benefit level is in significant, if any, jeopardy in light of negative investment experience.
Consequently, beneficiaries were not aware that such weakening in fact took place. Only actual events will tell whether the de facto guarantees will be withdrawn and, if so, to what extent.
One of the participants aware of the oft-quoted example of the Dutch pension solution (target benefit plan) reported that when actual pensions payments had to be reduced due to negative investment experience, the political backlash threw the future of this arrangement into doubt.

Risk Appetite

Plan sponsors are reluctant, maybe unable, to communicate in specific terms the level of risk they are comfortable with.
Reasons quoted included lack of incentive to do so (they are not the risk bearing party, and to some degree may be conflicted), transient staffing of this function and the lack of thorough understanding owing to the level of effort required to obtain such understanding.
Trustees usually operate with the understanding that they have to act solely in the beneficiaries’ interest; and this definition is broad (should tax payers be considered beneficiaries). Even without contemplating tax payers, beneficiaries’ interests are not clearly defined and can mean different things for different classes of beneficiaries.
The prevailing view is that while the level of risk taking should be the trustees’ responsibility, they are not in a position to set such level appropriately given the divergent interests involved and the lack of actual knowledge based on direct thorough communications with the parties about their preferences as to appropriate level of risk taking.
While originally risk appetite drove the level of risk taking, the current tacit view that has evolved in the past two decades is whatever is required to meet the funding discount rate is the actual risk appetite.
A further complication is that risk appetite is usually defined in investment terms such as standard deviation of return or VAR, and not in funding or benefit security terms.
While they represent close to 50% of all pension assets, under current practices, retirees bear no risk. Given changing demographics and limited payrolls, risk capacity should be of concern to both sponsors and active members, as well, thus a subject of serious study
For quantification of such risk appetite definition, one of the participants suggested a measure — assets/payroll—that looked like a good starting point in any risk appetite discussions.

Role of Pension Trustees

Time did not allow to discuss whether there is a difference in the role of trustees between pension funds only, pension plans (including jointly and not jointly trusteed arrangements). That probably should be the subject of a future Think Tank Session.
There was general agreement that trustees should communicate the level of the assumed risks and potential consequences to plan sponsors and the representatives of the beneficiaries. Participants then discussed how trustees can do that with effect and clarity.
It was suggested that the annual report can be used to good effect by not only generally discussing the need to take risks to improve plan returns, and the need of the expert management of such risk, but it also can explain who bears the risks and how it is ultimately borne by active members, retired members and/or taxpayers.
While current practices do not follow such communications standards, most participants were of the view that above suggestions would lead to much improved discussions on risk bearing, risk appetite and transparency without drastically changing how public pension plans run on a daily basis.
Over time, these changes, ultimately leading to a better understanding by the risk-bearing parties, will likely result in a push for a societally fairer and more transparent pension arrangements, such as target benefit pension plans.
Pursuant to the session, follow-up discussions with participants highlighted two critical provisos:
- Design features of target benefit plans, while ultimately fairer than a traditional DB pension plan, could be more complicated than straightforward DB plans, thus may pose even greater communication difficulties
- Such difficulties are not necessarily owing to complexity but sponsors’ reluctance – on account of electoral politics — to be clear about the probabilities of not meeting beneficiaries’ income expectations and the costs of their security.
If so, even radical changes in benefit design may not solve the “expectations problem” as the above quoted Dutch example had shown. Target benefit plans should be communicated in a way that directly addresses how benefits will be adjusted (in light of investment and longevity experience) thus removing expectations of a static benefit level.
Over time, these changes, however, may likely result in a fairer and more transparent pension arrangements, such as target benefit pension plans.
Trustees also have to ensure that objective measures enabling the assessment of appropriate risk taking are discussed and eventually developed by the sponsors.