In Depth: Unilever Case

December 14, 2001 ( - In the end, Unilever Superannuation Fund (USF)'s £130 million negligence suit against Merrill Lynch Investment Managers (MLIM) ended not with a bang, but a whimper.

Without admitting guilt, MLIM paid Unilever an undisclosed sum – estimated to be around £70 million – to settle the case that threatened to change the way asset managers and pension trustees do business. That conclusion leaves plan sponsors and fund managers alike wondering about its implications.

The case, which played out in Britain’s High Court, saw the consumer goods giant accuse MLIM of negligence in its management of Unilever’s £1 billion mandate. It specifically addressed the £600 million UK equity portion of the portfolio, managed by Alistair Lennard, a “wild card” in the words of Unilever’s lawyers.

Reports state that under the contract between MLIM, formerly Mercury Asset Management, and Unilever, the portfolio was required to outperform its benchmark by 1%, and lag it by no more than 3% for any four consecutive quarters.

In 1997 the portfolio returned 9.9%. While not losing any money for Unilever, Mercury failed to beat the benchmark, which increased by 17.8% over the year. In the first quarter of 1998 the fund returned 9.8%, against a benchmark return of 11.2%. Unilever fired Mercury. It claimed Mercury did not provide sufficient downside protection and enough diversification.

Target Practice

Many fund managers and plan sponsor might feel that this is a highly unusual set of performance targets. One portfolio manager told “Why set a target that gives more room for downside than upside?”

One risk control expert at a well-known asset management firm thinks otherwise, “Many of the most sophisticated investors simply don’t understand the measures used to track a portfolio’s performance. The 1% requirement refers to a return target, while the 3% refers to the tracking error,” he said.

He points out that the time-horizons set for the tracking error in many investment contracts is often mismatched, adding to the confusion. If a tracking error is set with the long-term time horizon in mind, and then measured over the short term, the numbers may very well appear out of range, he explains.

“In Unilever’s case the measure chosen might not have been the correct one, and perhaps the parties involved did not have an adequate understanding of these measures” he added.

Merrill’s lawyers argued that their was no mention of any downside tolerance targets in manuscripts of meetings of Unilever’s pension fund trustees. They questioned whether the issue of downside tolerance was regarded as significant enough to be discussed at these meetings.

Even Hugh Stirk, a former member of the USF Board of Trustees testified that there was a lack of clarity in the contractual obligation. The judge in the case also seemed to agree, suggesting that the targets were “nebulous.”

Reasons to Fire

Another portfolio manager points out that, "Investment guidelines generally quote performance measures such as exceeding the S&P 500 by a certain percentage, over a full market cycle."

"Sometimes the larger plans may stipulate that the investment manager must be in the top quartile or quintile of their peers' performance for the trailing number of years. However, not meeting these targets are reasons to fire managers, not sue them for mismanagement," he adds.

Many plan sponsors use a watch list system. Clare Murphy, executive director of the San Francisco Employees Retirement System with $10.9 billion in assets, explains that if her funds managers underperform for two quarters they are placed on a watch list and monitored very closely. "We terminate the contracts of managers who underperform."

Maryland State Treasurer Richard Dixon concurs, "We give our fund managers three years to perform. If they underperform, they're out."

But Unilever not only fired Mercury. Unilever sued.

Unsettling Questions

And then they settled. Fears that a precedent-setting win for Unilever would clear the way for pension trustees to sue their investment managers for underperformance have been put to rest. Or have they?

While a Unilever victory may well have drawn a clear-cut line in the sand, a settlement has muddied the waters for many. Pension funds that were clients of Mercury during the period in dispute, such as J. Sainsbury, AstraZeneca and Surrey county council, are now weighing their options, according to the Financial Times.

The settlement has ensured that the question of responsibility - and who should bear it - remains open. £70 million is a large chunk of change, and pension fund trustees may now feel pressured as fiduciaries to take their lackluster managers to court.

"While in the past, fund managers who underperformed their targets, were threatened with the termination of their contracts, now they're threatened with lawsuits," veteran fund manager John Remmert points out.

"It clearly puts fund managers on alert on how contracts and performance guidelines are written, and how performance targets are measured," he adds.

San Francisco's Murphy notes that her board would consider suing if a manager failed to comply with the contractual arrangements established and failed to make the trustees aware of his actions. "Managers can operate outside the guidelines if the market conditions warrant it. But they must alert the trustees," she adds.

Risky Business

The case turns the spotlight on risk controls. And begs the question - are pension fund trustees responsible for monitoring the risks taken by their investment firms? Certainly, Unilever's pension fund trustees are of the opinion that measuring risk is not the responsibility of a pension fund.

Wendy Mayall, the pension fund's chief investment officer, admitted to being unfamiliar with the concept of risk, according to the FT's coverage of the case.

In her testimony, she contended that she did not regard it as her job to tell fund managers how to do their job. She therefore did not scrutinize a crucial performance report, prepared by Mercury, too closely. Despite being asked by the fund to examine it, she conceded that she had only "cast her eye over it."

According to the FT, USF trustee Hugh Stirk concurred: pension fund trustees are not experts in monitoring performance. "It's not their job to measure day-to-day risk or decide how many stocks to buy."


Some plan sponsors would agree with these sentiments. Maryland's Dixon says: "We're not too concerned about risk measures. Performance is what it's about. It's what it has always been about."

Others take a view that falls somewhere in the middle, Murphy notes that, "plan sponsors are responsible for ensuring that their managers operate within the agreed upon investment guidelines. While it really is the manager's responsibility to ensure that the risks taken, fall within these bounds, plan sponsors should have some understanding of the risk control process, particularly with the more risky asset classes."

Stirk testified that the firm had a good track record, that the pension fund trustees believed the fund manager was in control, and were aware of Mercury's house style and its concentration on a limited number of UK equities.

They were not, however aware that investing in so few stocks would be so risky.

How much should pension fund trustees know about risk? "A lot more than they do," says Ethan Berman, CEO of Risk Metrics. "This case should put more pressure on plan sponsors to understand exactly how their money is managed, and that's a good thing."


Risk Savvy

To be sure, risk control at Mercury reportedly sometimes amounted to little more than a tap on the shoulder, a practice that left it open for Unilever's case.

The settlement might mean the end of that kind of 'flair-based fund management,' at least for the time being. The quantitative approach to risk management is back in vogue.

"The popularity of quantitative, versus flair-based fund management and risk control comes and goes, right now we're seeing an increase in the popularity of quantitative tools," one portfolio manager told

And even those quantitative tools are being closely examined. Some funds have moved away from using tracking error as a measure, towards a value-at-risk approach - a more comprehensive risk control system. It primarily focuses on downside risk and incorporates the value of the assets held.

"It's the amount of money at risk for a specific time period at a specific confidence level," Berman explains. "But by far the most important thing in terms of risk control is transparency and communication, which is not a number," he stresses. "You must know what your manager is doing with your money, and the way the information is conveyed is less important than that it IS conveyed."

Whichever approach wins out, pension fund trustees will no doubt examine these measures a little more closely. Some believe that fund managers will now emphasize the merits of their risk controls when pitching for business, and give pension trustees more information about these measures in their quarterly reports - which, unlike Mayall, trustees will make a point of actually reading.