The analysis and management of costs within a pension fund are grossly underestimated, wherethe area should be of the highest importance to managers.

This was the message given by Patrick Scheuer, director of the IBM Belgium pension fund at a recent European pension fund conference in Luxembourg.

If we are to respect the fundamental nature of a pension scheme, which is not to manage assets but to deliver a final service of pension provision through asset investment, then it is essential for a variety of reasons that the internal costs are carefully considered and accounted for," Scheuer asserted.

The principle difficulty in identifying costs, he conceded, was to be complete and exhaustive. Looking cross border at assets being managed, there are numerous costs hidden away in organisations, which are not easy to identify, he said.

"When it comes to making a decision on buying management services or indeed carrying out these services yourselves, if you do not know the exact costs involved then obviously you could fall into the trap of making an expensive wrong decision," he warned.

In terms of management fees, Scheuer's solution to this problem lay in an explanation of all related costs prior to the commencement of any arrangement between fund and manager. "All fees should be 100% defined before any agreement is reached, so the pension fund knows exactly the cost of the service it will receive," he stressed.

A major issue in the past, he noted, had been the rise in managerial fees proportionate to the size of the fund managed. This needed to be looked at in detail, with room made for econ-omies of scale and research expenses, but above all a revaluation of these spiralling costs.

He explained that his experience at IBM had shown asset managers to be open to this sort of renegotiation in order to find a "win-win" situation for both sides.

In choosing an asset manager though, Scheuer recognised that the issue was not solely one of cost, nor the question "do we take the cheapest?". "We should not be looking to follow the same managerial styles and portfolios as our peers, but taking time and asking for advice to balance cost and quality that suits our own mission," he argued.

Moving to the question of management performance, Scheuer advocated a similar predefinition of parameters before any concrete agreement. The setting of the traditional benchmark should be made without any room for interpretation and must be a 100% objective, he stated.

Above all, he stressed that any performance-related fee involved should not affect the key pillars of management style.

"What we have done at IBM is to stretch our performance interpretation over three years, and normally this spread will ensure that we avoid any risks involving asset managers," he continued.

Scheuer was certain that this performance spread, although not strongly welcomed in Belgium, was gaining popularity in the rest of Europe and would become steadily more common in the future, particularly in the use of large portfolios.

The key point of cost here, he suggested, was the link between the complexity of the investment mission and the actual fee being paid.

Again, in the field of transactions the importance of long-term strategies over short-term quick-profit tactics was emphasised. Regarding this, Scheuer conceded that the temptation might seem to tend towards index funds, but suggested that real value lay in determining consistent net performance over the three-year period, whichever investment methods were chosen. He also suggested that if the asset manager were independent from any transaction fees, then this would lead to more openness and objectivity.

Pension fund internal costs also represent a major issue. Scheuer asked why these were rarely identified and whether indeed it was worth doing so in terms of the overall cost picture.

Those he identified were the number of people working for the pension fund but not actually on the pension fund pay roll, including in-house consultants and lawyers, as well as IT operatives. These would affect costs if IBM were truly independent, he stated. "Identifying these costs is not to enable us to have perfect reporting, but more importantly to allow us to put on the table the basic reason for our existence in terms of our goals and the service we deliver to our employees. This allows for example the fundamental choice in Belgium of an employee between pension fund or life insurance."

Scheuer also identified the need for 'light' reporting responsibilities for a pension fund once an asset manager had been brought in. "We don't want the situation of having 300 lines, ie, sets of bonds or stocks, to one manager." He saw the way around this coming through the creation of ap-propriate Sicavs with asset managers, but also noted the advantages of pooled funds to share costs if this were not possible.

Similarly, he advocated the use of custodians, not only for the minimum physical holding of shares and bonds, but also in order to have an independent third party involvement in consolidated reporting and centralisation of accounts.

"We use this global custodianship at IBM to check the different values of our managers and speak with a unified voice when reporting them. This is of course a 'value added' mission though, and there is a price," he said.

Regarding taxation, Scheuer also stressed the need to be conscious of the kinds of Sicav vehicles being used and the tax levels of the countries these are invested in, as well as the existence of cross-border tax treaties. He used Luxembourg as an example of a country with an attractive fundamental tax scheme, but narrow bands of taxation that could imply additional costs.

The necessity of good-quality actuaries in a company was also highlighted by Scheuer, in terms of both their ability to anticipate and match fund commitments with investment returns, as well as avoiding problems with formal legal costs.

In conclusion, he stressed that cost should not be seen as an ancillary issue against return and risk, but as a necessity ensuring future safety and transparency of pension funds. Accordingly, cost "deserves its own monitoring tool if this visibility is to be achieved. We should constantly be challenging costs if the best decisions for funds are to be taken in their investment management and pension provision. The tools must fit the job." Hugh Wheelan"