Pension funds are like vulnerable old giants reaching with stiff fingers to fasten their skates before launching themselves onto the slippery ice of the future. Jean Frijns, chairman of the Frijns Committee, told us how he thinks these giants can be made fit for the future, Jean Frijns

"The task of the committee was to examine how pension funds' risk management and investment policies have developed and to look at the kinds of structural changes pension funds are dealing with now and in the near future," says Jean Frijns. "Our focus is economic; the norm against which we are measuring current practice is that which we would expect of a prudent expert in a given situation. We carried out a survey of 35 pension funds and then we held discussions with fund management heads and with external experts."

Recent losses were a reason for carrying out this study, but the aim is not to assign guilt, Frijns stresses. "As I said, we took a look at the structural developments and the challenges that they pose to pension funds and their management boards. We also looked at the past, but mostly to see where things can be improved. We're not in the business of naming and shaming.

"More concretely, we've tried to establish what economic best practice would be from the point of view of common practices in the sector. There is a lot of diversity among pension funds: there are some funds which are closed to best practice, while others are a long way from that."

Ageing giants
The biggest concern emerging from the analysis is the funds' vulnerability, Frijns says. "We use the image of ageing giants who, if things go badly, could end up in the position of sinking giants. Pension funds are being confronted with big risks: longevity; inflation and yield; market; all without mentioning execution risk. Set against these you have management instruments that are getting weaker: demographic ageing places a heavier burden on premiums and shortens the return period; premiums are high and can hardly rise any further; the company is cutting back on its supporting role; and low real interest rates offer few alternative sources of return. You'd expect that this greater vulnerability would lead to a strict risk policy and a clearer sharing out of risks among groups of stakeholders. That is already partially the case, albeit with huge differences between funds."

How far can the losses suffered by funds be put down to shortcomings in their investment policy and how far to external factors?

"It's hard to say. The main problem is that management boards are caught between two duties - between their real-terms targets and their short-term nominal constraints. There are pension funds whose investment policy is based on their real long-term aims, while others look at their short-term nominal coverage ratio risk."

It is striking that management boards rarely seem to set their policy on the basis of the actual status of their fund. "The committee has not succeeded in finding a clear link between the policy chosen and the fund's characteristics," says Frijns. "A complicating factor is that the pension contract and the risk framework are often very diffuse."

Dynamic, not active
In the process, the Frijns Committee has established that only a few pension funds are attempting to control downside tail risk via active management. "The question is whether boards are not placing too much trust in fixed allocation rules and stable long-term expectations. A dynamic policy - aimed at managing downside risk - appears to be the exception rather than the rule."

A dynamic policy, for example, would adjust allocations on the basis of changes in the cover ratio. The rules that determine when and how a policy is adjusted can be set up in advance. "The execution of a dynamic policy can be as strict and mechanical as you like. The decision on the degree to which protection is adjusted is not a mechanical decision, but we believe this decision should also depend on an assessment of the imbalances that are currently in place on the financial markets," Frijns explains.

And execution has not always been spotless. "This, too, is a case where there are marked differences between pension funds, since the survey shows that the differences in z-scores between pension funds are not particularly persistent." Furthermore, active management can look bad. "In bad years for markets, active policies can deliver a moderate negative contribution. That makes people assume that active policies are paired in practice with added investment risk."

Beyond this, during the credit crisis, problems arose not just for investments but with yield movements as well. Pension funds are obliged to use the swap curve as as a discount rate to calculate their pension obligations. This hit the floor during the crisis. The committee recommends using market yield as a starting point but not to adhere to it rigorously in all anomalous circumstances. But Frijns advises against using a long-term average yield. "Smoothing is too crude an instrument for filtering out erratic market movements; we are calling for a temporary freezing of discounting in times of market stress."

A small-scale study of pension funds carried out in Spetember by the DNB, the Dutch central bank, concluded that risk control in fund management was falling (seriously) short. Does Frijns sympathise with this conclusion?

"The committee was aware of the DNB study. It showed that some funds have made some serious mistakes in their execution. We have no reason to doubt these results."

The committee believes the explanation for these faults can be found a lack of managerial knowledge and expertise. "People came up with investment products that are complex, opaque and risky, almost certainly under pressure from external managers and investment banks." Of course, it is easy to say this with hindsight, Frijns adds. But nor has the committee shrunk away from its conclusions. "For the committee, the lessons learned were the first priority."

The first lesson is a need to strengthen management processes. "Our advice is targeted first of all at management boards. They must be in effective control - which means they must be in control of execution as well. That means managements need to have specific expertise in the fields of risk management and investment policy. In addition, the management has to have adequate processes in place so they can stay in charge of execution."

Even more important in the view of the committee is the role of management in setting a strategic risk profile. "That has to be done in a way that bears in mind risk exposure and the risk preferences of scheme members and possibly other stakeholders as well. A pension fund management board has the role of a trustee. The board has to manage the fund's assets in the best way it can in the interests of the participants and groups of participants. The trustee model in itself is very defensible; but the committee thinks it is high time that the paternalistic trustee model in place develops into a democratic trustee model where transparency and responsibility to participants is the first priority."