Currency overlay is a currency risk management technique whose popularity has grown in recent years. Five years ago there were only a dozen or so specialists offering the service – now there are probably more than 40 jumping on the bandwagon. So, for the treasurer who is concerned about which hedging strategy to adopt, or the asset manager who wants to know if he can use currency funds to add some diversification to his portfolio, at last there is some choice in the market. But what exactly is out there?
The issue of currency overlay provision has changed subtly. The pension fund manager or treasurer who has heard about the technique and wants to utilise it will no longer have the problem of finding someone who can provide the service – the question is, what kind of overlay does he want? Or perhaps a better question is, what kind of overlay will suit his business?
Overlay styles are many and varied. That so many of them thrive is indicative of the fact that the currency markets have many different aspects and that there are several theories of how they work. While it is probably fair to say that a number of different approaches ‘work’ in the sense that they reduce risk and can add return, they can vary greatly in ease of use and flexibility. The following list includes most of the styles commonly in use:

‘Top Trader’
This style uses a trader or more commonly a team of traders to make decisions about when to buy and sell the currencies, and which ones to hold positions in at any one time. Essentially it’s the outsourcing of the income stream from an experienced trading desk. For asset-management strategies, which just want a P/L stream with a good Sharpe’s ratio, this can work very well. However, it is not very easy to tailor to individual exposures, or for one-sided benchmarks. Additionally, the loss of a single individual can heavily impact the programme.

‘Economists Expect’
The last thing that economists often expect is to have their predictions ruthlessly tested. Often their job is more involved with retrospective explanations of currency moves than with hard cash placed on their forecasts. However, there are some courageous economist teams prepared to make predictions which can be used to place currency positions in the market, and thus to provide overlay strategies. These tend to be long term, low trading frequency strategies which are useful for similarly long term hedging strategies. Like ‘Top Trader’, it is sensitive to losses of critical personnel.

‘Super Statistics’
This type of overlay can be more demanding on the supply side, needing as it does both a trading team and a research group. The two collaborate in the model creation process, using ideas from both disciplines, and the research group then backtests the models using historical data. From this a pattern of expected returns may be obtained. If they are promising enough, a signal generation system will be developed by the research group and the model will be paper traded for a period of time. After both the traders and the researchers are satisfied that the model is acting as it should, actual trading may commence. This approach has the advantage that one model, if it is good, may work in several currencies. Additionally, models with different risk characteristics may be combined to suit a particular investor

Critical combinations
A number of overlay providers actually use combination methods, whereby they have a ‘stable’ of models which give signals, which the traders may then choose to follow or not. This is essentially an extension of the ‘Top Trader’ system, with all its advantages and drawbacks.

Track record and backtesting
It is difficult to decide between these different styles, but solid past performance is always a good indicator. Whatever the overlay style, a 10-year track record of successful trading will always, rightly, attract investors. However, without such a track record, it is difficult to make projected performance claims. The only overlay style which can be properly backtested using historical data is Super Statistics – because it does not deviate from tested models, it’s possible to generate expected returns and expected volatility of returns from past data. This kind of generated history is obviously not quite as good as an actual trading record, but it is a good indication of how the strategy may be expected to perform. Moreover, strategies may be designed for individual clients which may then be backtested to ensure that they provide the desired risk return profile. This can be difficult with other overlay types, though a certain amount of flexibility can be obtained by leverage techniques.

Risk appetite and benchmarking
Once a pension fund manager or corporate treasurer has familiarised himself with some overlay styles, he has then to think about benchmarks and what the risk appetite of the fund or the company should be. The benchmark will be the so-called ‘risk free’ position, which may be unhedged, fully hedged or some intermediate hedging level. It is important to define exactly what this benchmark is, as it is all too easy to assume that it is one thing when in fact it is another. for example, the pensions manager who is told that being fully forward hedged is ‘safe’. While this might appear to be a 100% hedged benchmark, if he is penalised for losing hedges when the currency moves in his favour, but when the market moves against him and he is unhedged then it is ‘bad luck’, then in fact he is being incentivised on a 0% hedged benchmark. While the majority of treasurers would describe themselves as having a fully hedged benchmark, in fact when examined most of them tend to have a benchmark which is close to zero hedged.
Risk appetite is another critical parameter when selecting an overlay strategy and provider. All hedging involves some type of risk – forwards risk losing hedges, options risk the loss of the premium. Currency overlay strategies have the risk that they will not perform as expected and will end up losing money. Here is where sound risk management techniques can play a part.
The pension fund manager should be shown actual or simulated returns for a number of years, for his precise strategy. Loss limits should be extensively discussed, and the following questions should be debated:
q Is the loss limit an absolute or a drawdown amount?
q Is it a periodic figure ($500,000 in one month, for example) or is it independent of time?
q Is it a guaranteed maximum loss? Most overlay companies are not able to guarantee a maximum loss, and if they do not, then their handling of a strategy close to its limit should be discussed in detail.
q Are stop losses set in the market?
q Has the overlay company ever gone through a loss limit before? If so, by how much?
With this set of facts under his belt, the pension fund manager will be able to decide whether he wishes to manage the whole of his exposure with overlay techniques, or whether only a proportion of it should be outsourced, leaving the rest to be forward hedged or left unhedged.

Choosing more than one
The questions of which overlay style and which overlay manager to select is not always answered by a simple choice. For decades investors have known that a portfolio is more than just the sum of its parts; that diversification means that return per unit risk is better for a collection of different investments than for any single one. The same is true of currency overlay. Choosing two or three managers with contrasting styles may yield dividends in terms of overall Sharpe’s ratio and returns. Another way of achieving a similar objective is to outsource just a proportion of the exposure, managing the rest internally. Indeed, a good overlay company ought to be able to design a complementary strategy to the current internal management strategy, as long as past data for the internal process is available. Overlay strategies need not be seen as competitors to the in-house traders – they can be useful counterparts.
Jessica James is London head of strategic risk management advisory, Bank One

Topics