Sections

Understanding currency overlay

Investment vehicles in looking for diversification of assets are increasingly searching beyond their domestic boundaries. This means they are
building currency risk into their portfolios, which has to be managed and reviewed. Currency overlay managers(COM) are able to manage this extra area of specialist risk. Their remit doesn't just end there, unless they're taking a passive approach. Active overlay managers are looking for an alpha source, in addition to fulfilling the functions of an overlay strategy.

We're in the business of understanding the market trends, and coming up with meaningful analysis, to help currency managers review their results, and plan more effectively for the future. Presently, very little analysis exists as to whether currency strategies are effective, and where they stand compared to other asset classes. Even less information exists on a sound, feasible methodology to tackle currency returns and risk. Our currency survey seeks to address these problems, while suggesting ways in which we can improve currency analytics.

We have just completed our third survey of COM, analysing data to December 2005. The inception date for the survey was August 1988.Previous surveys went to June 1999, and December 2003.

The foundation for our analysis is cemented in the frequently asked questions we seek to answer.

❑ Are different currencies correlated to each other?

❑ Can we differentiate between alpha and beta?

❑ Can we look at returns on a risk-adjusted basis?

❑ A number of different approaches are used to tackle the issues.

Initially, in the survey, we're looking at the structure.

We list the participating managers, and then show a breakdown between base currency and hedge ratio, using graphics and tables. The major currencies of USD, AUD, JPY, GBP and EUR are covered with hedge ratios of 0%, 50% and 100%.

Analysis is from account/manager participation levels, as opposed to capitalisation. From analysing 187 live accounts (December 2005), 50.3% are USD hedged, while 62.0% fall in 0% hedged vehicles. Since inception, 680 accounts have been included. This provides the reader with a top-level view of overlay managers, in terms of base exposures and hedge style. Composites are calculated covering all base currencies and hedge ratios.

The underlying methodology supporting our analysis has been excess returns. From these returns, we have provided an average annual gain per account. For the 2005 survey, the annual gain came to 83 basis points (bps), slightly lower than 2003 (97bps) and 1999 (106bps).

Average monthly gains/losses have been calculated per account/manager against the different currencies. The EUR accounts/managers showed the largest average monthly gains/losses. tracking error, excess return andiinformation ratio are measured. USD accounts/managers showed the largest excess returns at 0.73% and 1.29% respectively. EUR accounts/managers had the largest tracking errors, which confirms their volatile gains/losses record.

For the most recent survey, and to improve our risk coverage, an analysis of kurtosis and skewness has been introduced. Kurtosis measures the relative flatness of a distribution curve compared to a normal distribution. All accounts/managers have positive values, reflecting a peaked distribution. Skewness shows the degree of asymmetry of a distribution around its mean. Positive skewness was achieved for all currencies, other than JPY. Positive skewness indicates a distribution with an asymmetric tail extending toward more positive values. Negative skewness indicates a distribution with an asymmetric tail extending toward more negative values. USD accounts/managers had the highest positive values of skewness. Around and in support of this analysis, we built frequency distribution graphs for all currencies/composites, at manager level.

In the last survey, we added a total return methodology. The foundation for this approach was to take excess returns and add a standard benchmark return of US three-month Libor. This enables us to use total returns and use the principle of Modern Portfolio Theorem (MPT), which cannot be analysed fairly under excess returns. We accept there are limitations with this approach. A USD 0% hedged fund may enjoy a very different total return stream to a 100% EUR hedged fund, while sharing identical excess returns to their benchmark.

Therefore, applying a standard return of US three-month Libor+excess returns, to each fund, will generate identical risk/return statistics over a similar time period. This may contrast to using a true total return approach. We've taken the position relative currency returns are neutral, and a market neutral 0%, 50% and 100% hedge, may enjoy identical total returns since inception of the survey. We're concerned with researching alpha generation, and this approach still has merit, avoiding the inaccuracy of comparing different hedge ratios to an identical benchmark. We will look into building total returns at an individual account level, applying excess returns to the account benchmarks.

Building currency benchmarks is an area of much interest. At present, we are developing suitable benchmarks, accurately reflecting the style, currency, hedge and strategy of the underlying manager. This is quite a challenge. We see this as an area in need of greater coverage and accuracy.

 

otal return methodology reviews the different base currencies and calculates alpha, Sharpe and information ratios. Information ratio has been commonly used in reviewing currency returns, being the Holy Grail of risk/returns statistics for many analysts. Information ratio analysis is useful, but alone, ‘short changes' those looking to differentiate on a risk/return basis between different hedge ratios and base currencies. At account level, AUD had the highest Sharpe ratio of 4.39. USD and GBP accounts produced negative alpha. Risk analysis has been extended from introducing beta, then looking into upside and downside betas. GBP accounts had the highest value of beta and upside beta at 1.77 and 1.53 respectively. JPY accounts had a downside beta of -1.57.

Correlation is an important topic in currency discussions. For the accounts/managers, correlations have been calculated against a market proxy of US three-month Libor. Further into the survey, the individual currencies and composites are compared to each other and to indices from other asset classes.

The underlying strength of our approach to currency has been maintaining universes. These are compiled as follows:

❑ All currency all hedge all style;

❑ All Currency 0%, 50% or 100% Hedge Universe;

❑ USD/Non USD/specified currencies and specified hedge ratio;

❑ Specified style, currency and hedge ratio.

The universes are maintained on a quarterly basis, with software being developed, to produce monthly data. Universes can be exploited to produce quartiles, showing median, maximum and minimum returns. Risk/return scatters, distribution and value-added are produced also. Universe means are available to provide MPT statistics, on request. Universe results are included in the MAS currency survey. For the seven years to December 2005, the All Currency All Hedge Median return was 0.73%.

Analysis is undertaken between fundamental, dynamic and technical approach to currency management, looking at standard deviation, information ratios and returns.

Dynamic managers had the highest three-year rolling returns to December 2005, with 1.46%, compared to technical 0.32% and fundamental 0.74%.

The survey has introduced a new section dealing with currency alpha. We are collaborating with SEB in building a new index:SEB/Mellon Currency Alpha Index (available on Bloomberg). The rationale for this is:

❑ An unprecedented rise in interest from the institutional investor community;

❑ Active currency management is a genuine source of alpha;

❑ Separate out beta from alpha;

❑ Historical truth is that investing in the higher yielding currency and selling the lower would result in a profit being made;

❑ Overlay managers may be looking to reduce currency risk and not seeking alpha.

The survey is based on data from the COM universe, from which a quarterly report ‘currency overlay manager profile analysis' is produced.

Nick Rogers is product consultant at Mellon Analytical Services based in London

 

 

Have your say

You must sign in to make a comment

IPE QUEST

Your first step in manager selection...

IPE Quest is a manager search facility that connects institutional investors and asset managers.

  • QN-2466

    Asset class: Multi-asset funds.
    Asset region: Global.
    Size: EUR 30m.
    Closing date: 2018-08-16.

  • QN-2467

    Asset class: Search for a broker (mainly ETFs).
    Asset region: Global.
    Size: 250m.
    Closing date: 2018-08-28.

  • DS-2468

    Closing date: 2018-08-24.

Begin Your Search Here