For much of the past few decades, the retirement debate has been framed as a choice between two broad models:
- Defined benefit (DB) funds offered members greater certainty and pooled many retirement risks but placed considerable financial risk on the sponsoring employer.
- Defined contribution (DC) arrangements made contributions more predictable, benefits more portable, and gave individuals greater flexibility and choice.
But something else happened along the way. More of the risk moved to the member.
Today, a member of an individual DC arrangement may carry contribution, investment, inflation, sequencing, behavioural, longevity and retirement income risks. That raises an important question: have we become too comfortable with the idea that greater individual choice must mean greater individual risk?
At EPPF, we believe the answer deserves serious examination.
Members may want something between paternalism and complete individualisation. Choice matters because circumstances, financial needs and career paths differ. People change employers, experience interruptions to employment, support extended families and face financial shocks.
A modern retirement system therefore cannot simply recreate the rigid structures of the past. But neither should flexibility mean leaving individuals to manage risks that are inherently difficult to manage alone.
Our member research has been instructive. Members value flexibility but repeatedly express a desire for security and predictability. In EPPF’s recent member research, 61% prioritised predictable income over flexibility and control and, in one income-versus-lump-sum exercise, 81% preferred an income-based solution. That suggests the industry may be asking the wrong question.
It may not be: “Do members want security or flexibility?” It may instead be: “How do we give members meaningful flexibility without asking them to carry risks that can be managed more effectively collectively?”
The power of the pool
Pooling is one of the oldest and most powerful ideas in pensions. None of us knows how long we will live. Collectively, however, longevity is more predictable.
An individual investor may retire immediately before a severe market downturn, while a collective arrangement may spread investment experience across a wider population and longer period.
An individual may struggle to determine how much capital can safely be consumed each year over an uncertain lifetime. A collective arrangement can convert savings into income without requiring every member to solve the longevity equation personally.
Pooling does not eliminate risk. It changes where that risk sits and how it is shared.
That distinction matters. The future of retirement design may therefore lie less in choosing between DB and DC, and more in deciding which risks should remain individual, and which can be shared more efficiently.
CDC is one increasingly interesting answer
Collective Defined Contribution, or CDC, is attracting growing international interest because it occupies space between traditional DB and individual DC.
In a CDC arrangement, contributions are defined, rather than employers providing an open-ended guarantee of a fixed pension. But unlike individual DC, contributions and investments are managed collectively and retirement income is provided from the collective pool. Benefits are generally targeted rather than guaranteed and can be adjusted according to the financial health and experience of the arrangement.
In simple terms, CDC seeks to preserve some advantages of DC – particularly greater predictability of contributions – while retaining some advantages of collective pension provision, including longevity pooling and the ability to manage investment and retirement-income risk across a larger group.
The UK has moved from theory to implementation. The Royal Mail Collective Pension Plan is an authorised CDC arrangement, and the UK’s regulatory framework has been extended to provide for multi-employer CDC schemes. But that does not mean South Africa should simply import the UK model.
International systems make different design choices reflecting their labour markets, regulatory systems and social contracts. South Africa can learn from them and design for its own circumstances.
EPPF starts from a unique position
For EPPF, this is not an entirely theoretical conversation. Although EPPF is not a formal CDC fund, aspects of its existing architecture contain characteristics associated with collective risk-sharing.
Assets are managed collectively. Longevity risk is pooled. Investment and benefit outcomes are managed within a common fund framework. Members receive pension income rather than being left to construct that income individually, while certain member choices and transfer options remain available.
In other words, EPPF’s experience demonstrates that retirement design does not always fit neatly into a DB or DC box. The Fund has a long history of using collective mechanisms to help members manage difficult-to-manage risks.
That gives us an interesting starting point from which to ask what the next generation of pension design could look like. Could members have greater flexibility while retaining collective investment? Could some risks remain pooled while others become more individual? Could retirement-income security be strengthened without recreating an unlimited employer guarantee? Could collective structures improve portability, and digital tools give members greater agency without making them responsible for complex risks alone?
These are questions worth testing rather than assuming that the current architecture represents the end point of pension evolution.
CDC should not become another acronym in search of a problem
CDC involves difficult design questions. How should gains and losses be shared between generations? How much smoothing is appropriate? How should benefit adjustments work? What constitutes fairness between younger and older members? How much flexibility can be introduced before the value of collective pooling begins to erode? Will members understand what is promised, what is targeted and what can change?
International experience shows there is no single CDC design. Research and member testing should determine whether collective approaches can deliver a better balance of adequacy, security, flexibility and fairness for South African members.