Dutch regulators require pension funds to provide a 20-30% buffer on their nominal obligations. But who owns this buffer? The sponsor, the members or nobody?

For Theo Kocken, founder and chief executive officer of Cardano Risk Management in Rotterdam, the idea that the solvency buffers were ‘nobody's money' posed more questions than it answered. "When I started to look at the issue in 2001 I thought the answer to the question of who owns the buffer was the person who takes the risk. That poses the question of what is the value of that risk, and the only instrument that values risk is an option."

Kocken then picked up an idea that had been touched on by economists like Robert Merton and Tom Steenkamp, that pensions contain embedded options. Kocken used these concepts to calculate explicitly the real life values of these options. "A pension fund is actually nothing else than an accumulation of embedded options on top of individual savings pools," he says.

"Until now the presence and structure of these embedded options in pension funds have not been fully understood." More recently a number of pensions experts, including Kocken himself, Theo Nijman and Frank de Jong at the Netspar Research Institute plus practitioners like Niels Kortleve (PGGM) and Eduard Ponds (ABP), have taken a closer look at these options.

Kocken suggests that embedded options have resulted in some strange deals or ‘ curious contracts' between the stakeholders of pension funds.

And in his recently published book * he argues that the failure to recognise the nature of these contracts has - together with external developments like maturing of the pension funds and more transparent accounting rules - led to a situation where employers are dismantling their defined benefit (DB) schemes because they can no longer afford the risk they represent.

He also argues that a modified collective defined contribution (CDC) pension scheme could distribute pension fund risk more equitably and produce more sensible contracts between the different stakeholders than DB schemes or the current CDC solutions.

The first requirement is to understand and value the embedded options currently in pension funds, says Kocken. "During the last few decades, pension fund participants have agreed on various contingent claims to support stabilisation of the funding ratio and so add to the continuity of the defined benefit principle.

"Employers provided certain explicit or implicit guarantees to complete part of funding deficits, retirees have certain caps on inflation compensation and in the Netherlands and some other countries, indexation is contingent to the financial health of the pension fund.

"To make things even more complex, the employer can default on his debt and so on his guarantee to the pension fund, as the Pension Benefit Guarantee Corporation case in the US has painfully evidenced. This implies the beneficiaries have written an option on the joint default of the company and a deficit in the pension fund."

Kocken points out that the total value of these options can reach 30% or more of the pension fund's liability value - a risk that is borne by beneficiaries as well as employers.

Determining the value of these risks requires techniques in addition to the advanced techniques, such as scenario generation,that are applied in asset liability modelling (ALM), he says. "The existing techniques determine the aggregated risk for the entire pension fund, but they cannot determine the value of the risks in terms of what you should pay in the markets for guarantees similar to these embedded options. More important, they cannot determine which of the participants assumes what part of the risk."

Kocken's techniques enable pension plan designers to identify the different incentives for risk taking and the potential conflicts of interest these can create. An employer will want a pension fund to take less risk under certain circumstances, while a retiree will want it to take more risk in the same situation.

The techniques also identify conflicts of interest not just between employers and employees but between active and retired beneficiaries. "For example, younger participants under certain situations (for example when there is no employer to act as a guarantee provider), write options to retirees that can exceed 50% of total liability value. Since they are not explicitly compensated for these options, this is a kind of wealth transfer between generations," says Kocken.

It is important to identify the asymmetry of risk-bearing to understand why employers are currently transferring pension fund risk to their employees, as they move from DB to DC type pension schemes, says Kocken.

"Currently, the amounts of the risk that employers assume can easily take values of 25% of current liability value and even much more as a percentage of market capitalisation of that company."

Tougher regulations make the situation worse, he says "Restitutions, lower contributions and contribution holidays are now only allowed under very strict conditions. The bottom line is that employers face a situation where there is considerable ‘downside', far more than there was in the past with relatively smaller pension plans, a lower ratio of retirees versus actives and without fair value accounting rules like FRS17 and IAS 19, and little prospect of ‘upside.'"

If employers refuse to shoulder pension fund risks, what is the future of occupational pensions? Is a viable, sustainable pension system possible? Kocken believes it is. "A collective system is still viable without an active role by the employers. The Dutch Collective DC is already such a system, but adjustments are necessary to make it appealing long term to both young and old participants. In particular, the people who assume most of the risks - the younger actives - must be compensated explicitly when returns are favourable. That is where the current system is failing."

Kocken says that the younger active members of a pension scheme have a supply of ‘human capital' which makes them better able to absorb the shock over their long career horizon.

"A viable pension is not only about expecting equity markets to outperform in the longer term and then all will be fine," he warns. "It's more to do with the younger generation having more human capital. This means that they can be more flexible with contributions and they can cope with the risk. They can still take equity risk and in the long run they will have the additional equity risk premium in return. But if things go wrong, they can absorb large market shocks much better than retirees. However, they need a fair compensation for the risks they take, and this feature is currently absent in pension contracts."

The solution, he says, is a pension design which takes the features of the ‘life cycle' DC approach advocated by Keith Ambachtsheer and others and adds a collective wrapper. "The feasibility of such a scheme is not very different from a life cycle DC approach but it has the added advantage that it enables inflation-indexed pensions, which are not available or hedgeable in large volumes in current financial markets," he says.

 

Another important advantage of the new pension design is that no additional solvency buffers are necessary on top of the pensions savings. Currently the regulators in Holland and Scandinavia insist on a buffer of roughly 20% to 30% in addition to pension obligations.

Kocken argues that this is unnecessary and economically inefficient, creating pools of ‘dead money'. It is also pro-cyclical in the sense that when markets move down, all employers are compelled to pay out additional money.

In a system where only beneficiaries are stakeholders - such as a collective DC scheme - there is no need for such buffers, Kocken says. "In a system where only beneficiaries participate, it is pointless for the regulator to require additional money to protect the beneficiaries' own savings. A buffer only makes sense when someone else, for example the employer, is providing the guarantee to the pension."

Instead the active members provide the buffer for the retired members' savings when necessary.

"If things go wrong in the markets, the actives will have to pay something more. So for example if the value of the actives' saving money is 30% , as a percentage of total pension wealth, this serves effectively as a buffer for the 70% value of the retirees' pension liabilities."

This is economically efficient, as well as socially equitable, Kocken says. "The virtue of this is that it does not only rely on the long term outperformance of the equity market. It relies on the fact that the actives are more flexible in providing security because they can afford to pay say 2% or 3% more annually over a very long time horizon. It will also make the whole pension system anti-cyclical."

Kocken suggest that the pension system in Holland and the UK has now reached a tipping point and DB schemes are on the point of collapse. "It is now or never whether they will collapse in the right or wrong direction," he warns.

*'Curious Contracts: Pension Fund Redesign for the Future' is published by Tutein Nolthenius and also available via www.cardano.com