The euro rise has focused attention on the importance of currency management. If you are invested in non-European equities, and hedged the currency risk, then your fund lost 20.9% in the first half of this year. If your international equity managers did not currency hedge, then this loss was increased by a further 10.1% from the euro rise. Currency risk needs to be managed.
Currency management is a specialist activity: there are some 20 firms who offer this. There is a track record of over a decade, and active currency management is now employed by large funds around the world. Historically many European funds ignored currency. There were three reasons for this. First, international allocations were relatively low. Second, equity managers generally argued that “Currency washes out in the long run”. (Being older, and remembering the yen at 360 to the dollar, I disagree with this statement.) Finally currency management track records were short. The view on active currency management is now changing rapidly: the past two years have seen the most rapid rise in currency management since the activity took off in the late 1980s.
Value is added and risk reduced
The reason for growth in active currency management is not hard to find. Go around the world and meet with the leading pension funds. The topic of discussion is always the same: the need to improve returns to the risk budget. Having unmanaged foreign currency exposure just increases fund risk – with no associated expected return. If currency is actively managed, then value is added and a pre-existing risk is reduced. The industry average value added from active currency management was 1.1% pa over the past six years.
The key points: Currency overlay provides pure incremental return. Total foreign currency risk of a fund can be halved by active currency management, and value has been added. For funds who are building their international allocation, currency management is vital to secure the base currency value: You do not want to double the international allocation – and then have it wiped out by an avoidable, and unmanaged, dollar fall.
Even those who believe that currency returns ‘wash out’ are now turning to active currency management. Chart 1 shows the cumulative currency impact for euro based investors investing in ex-European equities. Euro-based investors enjoyed a massive windfall currency gain of 38.3% over the 24 months to October 2000. This was partly caused by – and amplified – the major shift into international equity investments by the large funded European pension plans.
Funds do not want to see all this currency gain ‘wash out’: from the peak in 2000. The currency fall from the peak through June 2002 has already been 19%. Interestingly, this sequence of events is exactly what happened in the US in the late 1980s. Sophisticated funds, such as the General Motors pension plan, had built their non-US international investments in the early 1980s. As the dollar fell from 1985, these funds enjoyed similar very large windfall foreign currency gains. So at the end of the 1980s, the US funds with large international allocations put in place active currency management. That was what really accelerated active currency management for institutional investors in the US.
Prospective returns are lower
The real driver for active currency management is the need to improve returns to the fund risk budget. Historically the answer to increasing returns was simple – just increase the equity allocation. Today there is no easy answer. Even after the stunning equity market falls, leading experts remain very cautious about long term equity outlook. Informed estimates of the equity return premium over bonds are as low as 2% pa.
With such a low equity return outlook, the average incremental return of 1.1% pa from active currency management becomes even more attractive. The overall return outlook is lower, and there is nothing funds can do about this. This is forcing funds to focus on the risk side of the investment equation. And that is the driving force behind the aggressive diversification that is occurring in all funds.
Diversification to reduce risk is
Many funds are exploring alternatives, such as private equity or hedge fund allocations. They are diversifying to reduce total fund risk. However these new initiatives typically have only a 5% initial allocation. Arithmetically, this can have little impact on total fund return to risk. The only action that can materially impact total fund return to risk is letting go of home country bias: hence international diversification is accelerating around the world. A 20% international equity allocation is the minimum. However the global trend is toward much higher international allocations, particularly in smaller open economies.
Historically, funds concentrated over 85% of their assets in their domestic markets. There were three reasons for this. First there were regulatory, currency matching, restrictions on international investments. Then there was peer relative performance pressure. Finally, there was concern over currency risk.
International allocations are
Regulatory restrictions have been lifted in most countries. Peer pressure has evaporated. Funds have rightly switched from measuring risk against their industry peers, to measuring their own funding risk against their own future liabilities. New accounting standards are a big factor in this change. The UK accounting standard, FRS 17, went too far in marking to market long-term pension liabilities. However, lower bond rates do imply much higher future funding needs. The move to IAS-based accounting means closer focus on real pension funding needs. Not doing active currency hedging creates large future funding risk.
Concern over currency risk is the only remaining constraint on raising the international allocation. Once currency concern is gone, international allocations rise very rapidly – for example, post the introduction of the euro, Dutch funds now see domestic investment as being European investment. Similarly, once active currency management is in place, we see funds accelerating the growth in their international allocations. For the same risk to the total fund surplus, the international allocation can be doubled once active risk controlled currency overlay is in place.
Actively managing the other half
of International equity returns
The initial concern over the impact of currency risk was, half, correct. If currency risk is not actively managed, then the currency risk certainly goes unrewarded. Consider the returns for a euro based fund investing in international equities. There are two halves to the return. First, there is the risk from investing the capital in companies around the world. There is centuries- long data to show that investing capital in equities is passively rewarded. Over the period January 1987–June 2002, hedged international equites provided a 5.7% pa return for euro-based investors. There is risk of loss, but the long-term return is strongly positive. Chart 2 shows the monthly distribution of equity returns over this period.
Second, there is the currency exposure. This is the currency translation swings that accompany the international equity investment. No additional capital was invested to create the currency exposure, so the expected currency return must certainly be zero. The average return over this period happened to be plus 0.8% pa. This is statistically not different from zero – exactly what theory suggests.
But zero expected return does not mean that the cumulative impact over any three year period will be zero. Stay with passive currency exposure and you are simply going to face a widening range of outcomes. You can see that fact in the 38% cumulative gain in the 24 months referred to earlier. And the 19% currency loss since that peak.
US institutional investors experienced the flipside of these moves: Not only did US equities outperform overseas equities in local market terms in the earlier period, the dollar strength wiped over 25% off the value of these international assets. Not surprisingly, new US allocations to international stopped dead in their tracks in this period. Now, with falling US equities and the falling dollar, US allocations to international are set to rise very sharply.
Note the sheer size and unpredictability of these currency moves. This has fuelled rapid growth in separate active currency management. Funds do not want huge multi-year currency swings as an unmanaged side effect of international investing. Funds are opting for specialised active currency management to turn the pre-existing currency risk into return.
Be clear that in active currency overlay management there are two goals. It is not just about seeking return. Of course, the only reason for appointing any active manager is to add value. But the primary policy goal is to reduce the risk of currency loss – which the fund already has. The only question in evaluating active currency management is the information ratio: how much active risk was spent by the fund to gain the incremental return. With a decade of experience, it is now possible to evaluate results.
Survival, reporting and
Five years ago, many funds were justifiably dubious of exaggerated return claims from some currency managers. The currency manager universe was small. Hence survival bias, reporting bias and emergence bias were major issues in analysing performance. For example, the first survey of overlay manager returns showed average value added of 1.9% pa. (This was the survey undertaken for BellSouth pension fund by Brian Strange of Currency Analytics.) However, this was a survey only of the managers who chose to reply (survival bias) – and the numbers they chose to give (reporting bias). We sent in the numbers on all our accounts, and thus accounted for over 20% of all the numbers in this survey. Several firms who had already left currency management were of course not included.
Emergence bias is another, more subtle, issue. In the late eighties there was demand for currency managers. So several global bond managers analysed the currency returns within their global bond accounts. The half with good currency attribution numbers declared themselves to be currency managers. They ‘emerged’ with good long run historic currency track records – extracted from their global bond accounts. At least three of these global bond managers subsequently left currency management. Their high risk, negative return numbers are not in anyone’s performance universe.
Surveys show active currency
Today these performance measurement biases are lower. This is because the surveys have longer track records and are more inclusive. The largest survey of currency management is by Russell/Mellon CAPS. This survey covers some $80bn in active currency overlay programmes. Note that the survey covers third party active currency overlay programs. Any global equity, or global bond, or GAA manager can claim to be a currency manager for the underlying assets they managed. (Of course, they only make this claim if the currency results have been positive!)
One of the surprises in the survey was the different active risk levels taken by different currency managers. Many managers take low risk (under 2% per annum tracking error), others take risk of over 3% per annum. Obviously managers taking low risk are appointed to larger, core, mandates. Here are the numbers for the six-year value added against the unhedged and 50% hedged benchmarks.
Note two caveats. First, currency returns are highly episodic. It is a fact that currencies spend much of the time moving sideways in a trendless fashion. Then there are brief episodes of large moves – such as we have seen with the euro in the first half of this year. It is in these environments of large currency moves, that currency management is needed – and in which there is the real opportunity to add value.
A second observation is that the manager universe returns for currency overlay were much lower over the past three years. This is because the three years to 2002 was a very adverse environment for some of the value-based currency market timing styles: they took large long euro positions - betting prematurely on a large dollar fall. In the first half of this year it is likely that these value-based currency managers had strong returns as the euro returned to par with the dollar.
What the active manager return table does not show is the reduction in pre-existing fund currency risk from the overlay. Funds engage in currency overlay because they are concerned about their total currency risk. So, while it is nice to know that active currency management adds value, it is vital to ensure that the fund’s total currency risk is reduced. The histogram in chart 4 shows the monthly unmanaged currency swings for a large euro-based investor in international equities.
The downside currency volatility on the international equity portfolio over this period was 7.6% pa. Currency fluctuations do not correlate in a stable way with anything. So, as European investors know well this year – currency volatility simply adds to the existing international equity investment volatility. And it is the downside volatility which we do not like.
Here are the results after active currency management. Forget about the return for the moment, the downside currency risk has been more than halved to 3% pa. The international equity asset class has been made safer: the allocation can be safely increased.
The very rapid growth in active currency management is no surprise: Value has been added, and successful managers have achieved this at under a 2% active risk level. Active currency hedging can halve the pre-existing currency risk that funds already had.
Active currency management ensures that accountable management for this major risk is in place. Finally the added value from currency overlay required no separate capital allocation by the fund: the returns were pure incremental returns.
So actively managing currency not only reduces the denominator in the return to risk equation – it raises return as well. For those funds who are looking to increase their international allocation sharply, active currency management is key.
Selecting currency managers
The policy question of when to start active currency management can absorb a lot of staff time at a fund. Hence the decision to use active currency management is often postponed until international is over 20% and/or a currency loss has occurred. Generally the currency decision coincides with a strategic review of the fund’s total asset, or asset/liability, position.
The task of selecting an active currency manager is easier. Many of the issues are the same as hiring any active manager. These are questions such as length of track record, depth of resources, backup facilities, stability of team, organisation and philosophy etc. However there are several special issues that are unique to selecting currency managers.
Six special issues
1. Request audited returns
It is important to ensure that there are independently audited, preferably AIMR compliant, live track records available. It is also important to ensure that the team responsible for the track record is still there. As there are so few skilled currency experts, there has been much staff circulation within the small currency management community.
2. Compare like with like
Performance results will not be comparable if the currency base or currency benchmark is different. For example, for five years until 2000, it was impossible for an active currency hedger to beat an unhedged euro benchmark for a European fund (although this year has given a huge value adding opportunity). Equally it was impossible for an active currency manager not to have added value if the dollar benchmark was unhedged over this period. This is why it is useful to examine track records from dollar, euro and yen currency bases. One can then ensure that results were not just due to luck, in one period, for one currency base.
3. Does the approach still work?
There clearly has been very major structural change in the currency market since the arrival of the euro. Many managers were able to add a lot of value prior to the creation of the euro, by exploiting European currency convergence trades. Some managers have found it difficult to add value after this structural change.
4. Is the activity currency hedging?
Some of the early claims of very high returns came from currency managers who were aggressively taking cross rate positions. Note that some currency managers argue that this activity is “hedging”. Increasing pre-existing currency exposure is not hedging under the conventional accounting definition1. It is not incorrect to do this: it does increase the manager’s opportunity set to add value. However it also increases the risk of large unexpected loss. We caution funds to start with strict hedging unless they have a sophisticated board, and experience with asset class activities.
5. Be aware of potential conflicts of interest
One has to be aware of potential conflicts of interest in trading. Currency is – uniquely – a dual capacity market: banks act both as agents and principals. For example, some managers aggregate currency trades across all clients – and then parcel the trades out. Currency as an asset class has fees of 100bp pa and 20% of profits. Whilst currency overlay fees go down from 25bp pa. There have been unfortunate incidents of currency trades being mis-allocated, and ‘commissions’ being charged for currency trading. These incidents were not in mainstream overlay firms. However, where firms aggregate trades and take profit shares, or take proprietary currency positions (in either the same or affiliated entities), extreme care should be taken to ensure total transparency of dealing.
6. Talk to other funds
Finally, it is very useful to discuss with other funds who have actually had multi-year experience with the currency overlay managers considered.
There are different active currency styles available. These range from risk based styles through fundamentally based timing styles, to technical styles and judgmental styles. Whichever way the active currency management is done, the evaluation is simple: How much value was added relative to the active risk taken?
This is an important point: You do not give a currency manager fund capital, you give them an active risk budget. This is a risk budget of contingent capital. Currency managers can only take currency positions, and thus add value, because the fund stands ready to meet any potential currency losses. Firms who have left currency management employed high risk, market timing styles – usually with judgmental directional overrides.
Hence it is important to know how much active risk the manager is taking for the fund. The simplest way to check this is to talk to their clients. Check the worst case live, not simulated, outcomes. Obviously higher risk approaches promise higher return: Do you want a promise of 2% per annum return, with the risk of a worst case one year loss of over 6%. Or a promise of 1% per annum return, with a worst case one year loss of under 3%? These are the sort of policy issues that must be addressed in selecting the active currency manager.
Currency management – you cannot leave home without it
Currency management is a complex issue. The topic is shrouded in controversies. Hence it takes time to put an active currency programme in place. This is why it was the larger funds, with internal staff resources, who moved ahead first to benefit from active currency management.
However, there is fundamentally no way of escaping the currency issue. Consider this: if your fund has a significant international allocation, would you ever take such large, totally unmanaged, currency exposures without the underlying equity investment? We have never met any fund that takes unmanaged currency exposure by itself. We know many funds who have this unmanaged currency risk as the side effect of their international allocation. They only accept this because they of equity managers’ assurances that “currency washes out in the long run”.
With today’s funding pressure, there will be no long run if the unmanaged currency loss is very large. The euro fell close to 80 cents against the dollar. It could clearly rise to 1.20. We do not know whether this will happen, neither does anyone else. But it is a risk. And it is the role of plan sponsors to identify and manage risks. You can choose to leave this large risk unmanaged. But the numbers show that active management of currency risk can halve this risk and add value. That is why you should not leave your home currency, in which your liabilities are denominated, without having active currency management in place.
Ronald Liesching is director of research at Pareto Partners in London