The currency turmoil after Thailand's devaluation last year made it brutally clear that pension funds cannot afford to leave their currency risk unmanaged. In the three to five years over which pension funds measure results, currencies can critically impact a fund's investment returns.

Although the euro will eliminate some currency risk, European funds with cross-border investments will remain exposed to large movements in the dollar and the yen. The dollar's movements since 1988, for example, can be divided into seven periods in which it rose or fell 27% on average (see chart).

Passive hedging removes currency fluctuations from investment returns. However, it really only moves the volatility to the cash that supports a passive hedging programme. In years when currencies fall, hedging gains accumulate cash, while in years when they rise, cash is lost when hedging losses are settled. Since currencies have a history of trending for two to three years, passive hedging is impractical and costly, as is passive hedging with options.

A Study by Currency Performance Analytics has demonstrated that currency risk can be managed. Active hedging reduces risk while it can add significant incremental return over time. The lack of studies demonstrating that active currency management works is one reason European funds have been reluctant to separate currency management from underlying equity and fixed income management processes. However, it is the first step to set up a currency management programme. Since currencies and equities often move in opposite directions, hedging decisions must be made independently. Hence the term currency overlay.

Sophisticated pension funds have realised that it is unrealistic to expect equity managers to simultaneously produce high equity and high currency returns. As a result, they began to appoint overlay managers in the early 1990s. Their initial aim was to reduce risk. However, as evidence accumulated that active hedging adds incremental returns, they have be-come increasingly return-oriented.

The second step is to select an overlay manager. Unfortunately, currency overlay returns are not directly comparable due to differences in underlying exposures and benchmarks. As a result, overlay managers cannot be ranked to identify superior skills. Even if it was possible, it is prudent to assemble a team of currency managers.

By investing in many industries and securities, equity risk in international portfolios is reduc-ed. However, the associated currency exposures are not diversified. A global equity portfolio with the euro as its base currency usually has three major exposures; the dollar (52%), the yen (10%) and sterling (11%) (see chart). Few plan sponsors would permit its funds to be invested in only three companies. It would be too risky.

If currency risk could be diversified with cross-hedging in other currencies, it would be reasonable to appoint one overlay manager. However, since the risk cannot be diversified, and any overlay manager may at times perform less well, it's prudent to reduce the currency risk through a team of overlay managers withdifferent decision styles.

The third step is to develop a mix of managers with the styles that are sufficiently different to ensure that manager diversification is achieved. It substitutes for the inability to diversify concentrated currency risk with other currencies. Although styles can overlap, they can be grouped into fundamentally-based, model-based and dynamically-based programmemes.

When the overlay team has been assembled, there are operational issues to consider. Trading lines must be arranged with several banks to reduce credit and delivery risks and to prevent one bank from taking advantage of 'captured' business. A reporting system must also be established between a fund's manager(s) or custodian and the overlay team to permit timely communication of changes in exposures to avoid over or under hedging.

An overlay programmeme's benefits must be measured over time. A plan sponsor usually selects a currency benchmark that reflects its risk and/or asset allocation preferences. It can range from 0% to 100% hedged. Regardless of its hedge ratio, it must reflect how underlying exposures can change over time. A ratio of 50% hedged has become generally accepted in the US. It is seen as 'market neutral' with no preconceived view of whether currencies will rise or fall. It also stimulates overlay managers to more actively move away from the benchmark to focus more on adding return than merely reducing risk.

The success that leading pension funds in Holland and Belgium are having with their programmemes and the accumulating evidence that currency management adds return, will likely lead to a rapid acceptance of currency overlay in Europe. The end to the dollar's rise since 1995 will again highlight, perhaps 'brutally,' that European pension funds need currency overlay programmemes sooner rather than later.

Alfred G Bisset is president of AG Bisset & Co in Connecticut