Delivering the performance
Most investors recognise that currencies have a large impact on the returns of international and global portfolios. By selecting a multi-currency benchmark, the investor, implicitly or explicitly, makes decisions on a set of underlying assets and on the desired level of embedded currency exposure.
There is a need to analyse the effects of currency movements and currency decisions separately from the underlying assets. Currency is a large source of return and risk, both in terms of benchmark selection and active management. This currency risk needs to be managed and specialist firms can provide currency management as a separate service, often referred to as currency overlay, from the underlying asset management. This has become well established over the last decade and provides the opportunity for risk management and added value.
Attribution methods now allow asset selection skill to be differentiated from currency selection skill. Currency performance and attribution results also highlight the significance of the choice between an unhedged, a partially hedged, or a fully hedged benchmark.
The change in spot exchange rate is often used by investors as a measure of the influence of currency on their portfolios. However, this approach can be misleading since it fails to reflect the actual returns that can be obtained by currency instruments and ignores the effect of the forward currency premium. We get better information about the true effect of currency if we split the return derived from the change in spot rates into two separate components:
The forward premium, which is known in advance and is driven by short-term interest rate differentials. The component of the change in spot rate not accounted for by the forward premium and which is commonly called currency surprise.
Investors cannot eliminate currency effects entirely and earn the local market return. They can only eliminate the currency surprise component while retaining the forward premium component.
An excellent example of the currency surprise phenomenon is provided if you imagine that you were a sterling investor reviewing the performance of the pound (£) against the Deutsche mark (DM) from 1972 to 1997. At first glance the pound was weak over this period since £100 converted into DM in 1972 would have been worth £253 at the end of 1997 through the pound’s spot depreciation. However, if the investor had sold DM forward for pounds, the total investment would be worth £340. This is because the cumulative forward premium was greater than the pound’s actual depreciation. In fact, the currency surprise of the Deutsche Mark was negative by £87, and the hedged return outperformed the unhedged return.
Because of the large potential differences in the hedged and unhedged returns and constraints on the manager’s ability to add value long-term assessment is fundamental. A currency overlay manager with an unhedged benchmark can only add value from hedging a currency when it is weak and similarly with a hedged benchmark can only add value from lifting the hedge when the currency is strong.
So when assessing whether active managers can add value the impact of a currency’s movement over that period is very important.
Currency overlay mandates
A currency overlay strategy is broadly one in which the management of the currency is carried out separately from the remainder of the assets. This could be carried out by the same asset management organisation or by a manager totally independent of the management of the rest of the assets.
Currency overlay mandates can either be active or passive. Passive currency overlay usually takes one of two forms. This can either be neutralising the currency effect (ie, the currency surprise) brought about from active country positions when managing against an unhedged benchmark. Alternatively a fully hedged passive overlay means eliminating all of the currency surprise.
Active currency overlay mandates seek to add value by participating in upside currency gain while protecting against downside currency losses. It needs to borne in mind that because the forward premium cannot be hedged away, hedging an asset will give you the hedged return (local plus premium) not the local return. This is fundamental in the understanding of these strategies.
We can think of the overall account return as consisting of three components broken down as follows:
o Hedged return – the hedged return on the underlying portfolio
o Surprise component – the impact of the currency surprise component across the whole portfolio
o Overlay return – the impact of the overlay strategy
As the benchmark can also either be fully or partially hedged, this breakdown enables us to distinguish between different components of the strategy and define benchmarks accordingly. The aim in selecting an appropriate currency overlay benchmark is that it must reflect the nature of the assignment. Since currency overlay is typically managed separately from the underlying foreign assets, performance should be correctly attributed by portfolio or manager and it is important to identify which components of the assignments are the direct responsibility of the currency overlay manager. In the survey results that follow we simply look at the added value the overlay manager delivers. These will be calculated in different ways depending the nature of the individual mandates.
Russell/Mellon conducts a survey of active currency overlay managers on a regular quarterly basis and the results now extend back over seven years. The survey currently analyses the results of 149 separately managed accounts by 19 asset managers, covering $54bn (E48bn) of assets. Accounts are only included where at least three major currencies are managed to add value. The overall universe captures the results from a range of accounts with different base currencies, different degrees of hedging in the benchmark and different management styles and so a number of sub-universes are also calculated to focus on a particular emphasis. The latest available results to 31 March 2003 are summarised in the table.
The results have been very good over the last year with excess returns generated in three of the four quarters analysed. Results tail off over the three and five year periods but remain firmly in positive territory and pick-up again over the seven year period due to very good results in the first two-years of the survey.
The overall results hide individual manager performance and clearly a range of results is achieved. However around three-quarters of individual manager results were positive across the periods shown. In addition the levels of risk vary by manager, the largest risk (measured as tracking error) at 3.0 to 3.5% per annum and the median between 1.5% and 2.0% per annum.
In the search for added value, particularly with prospect of a low return environment, currency overlay management has delivered in the past while managing the funds overall currency risk profile.
Alan Wilcock is research & development director for Russell/ Mellon CAPS in Leeds