Manager Writes: Currency Hedge

More than just the carry trade

ADRIAN OWENS and MARK DRAGTEN, AUGUSTUS ASSET MANAGERS
Originally published in the September 2007 issue

2007 has witnessed more large inflows into hedge funds with much of the new monies coming into long short equity/related strategies. Rising Libor in many developed markets (ie. US, Europe and UK) has increased the challenge facing fixed income hedge funds to produce returns that – net of management fees (typically in excess of 1.25% per annum) – are attractive.

Currency returns have perhaps been less attractive over the past few years. 2006 was a challenging year for pure FX returns, and with some exchange rates caught in a narrow range for most of the year there were not widespread opportunities for alpha generation in FX. According to a new report from Deutsche Bank AG, investors should significantly increase their exposure to currencies in order to stabilise and maximise their portfolio's returns. More than twenty years of data showed that average annual returns from pure currency investment strategies are around 10%. These returns are in excess of bonds over the same period and about the same as equities. Indeed, with scenarios of rising inflation and monetary policy shifts there should be increasing opportunities to break out of the relatively trend-less markets of the last 12 months.

Looking at currency-only hedge funds it would seem that much has centred on the benefits and pitfalls of the 'carry' trade. Merrill Lynch estimate that the total amount of FX capital pursuing carry trade ideas is about US$1 trillion, while the International Monetary Fund's April 2007 Global Financial Stability report states that there is about US$170 billion in the Yen carry trade alone. For those heavily invested in carry trades within foreign exchange the experience of Q1 2007 would indeed have been chastening. It may not be possible for currency hedge fund investors to avoid completely the impact that the unwinding of major carry trades has on their portfolios.

Our recently launched fund combines discretionary management and systematic models to give investors attractive risk adjusted returns that are derived from much more than the carry trade. Diversification is achieved through process, different trade durations on the strategies, a broad country focus, money management and products. Such a product combination can help to minimise the impact of adverse moves felt by those who rely too heavily on any one return source, most notably recently including carry.

We have been actively managing currency since 1983. The currency hedge fund approach splits the portfolio capital 50:50 between the discretionary and systematic approaches. However, the risk (VaR) allocation can be more skewed to reflect the portfolio manager that has the greatest conviction at any time in a particular set of currency themes.

Good discretionary management has the benefit of being able to shift focus quickly as market paradigms change. Augustus prefer to focus most of their discretionary currency risk between the G3 and G20 where they believe fundamentals are often clearer drivers of currency moves. One of the characteristics of their currency investing over the last 23 years is the relative lack of concentration in the extremely liquid and more efficient G3 (ie. US, Euro, Yen) markets. In these markets, currencies can also get pushed around by central bank reserve managers.

Over the medium-term we would argue that macro fundamentals such as economic growth, interest rates, inflation, current accounts and foreign direct investment (FDI) primarily drive currencies. However, they recognise that for long periods fundamentals can be completely ignored by markets and currencies are driven by other factors such as sentiment and positioning. Identifying these shifting market paradigms is key to successful currency management.

Good discretionary managers can often steal a lead over systematic approaches by predicting the shift from one key driver to another. An example of this would be the Czech Koruna. The country's economy has witnessed a lot of FDI recently, which has, for example, supported the modernisation of industry. However, a lot of the profits are now being repatriated and there is a good case for the Koruna to start trending weaker as capital flows become more balanced and with both inflation and interest rates low, the latter at a mere 2.75%. There is therefore a case for shorting the koruna against the high yielding Turkish Lira. Using options is arguably the most prudent way of expressing such a view given the associated tail risk from a carry wash-out. We currently have such a position expressed through call spreads and dual at-expiry options. In the event of an external shock or a currency unwind, the investor will lose no more than their premium invested. The potential return to premium paid on this trade, as structured at inception, was an attractive 9.5: 1.

For the systematic approach, where algorithmic rules-based models decide on the trades, the models can survey more markets more quickly than the human eye while maintaining the tested disciplined process. We currently run three systematic models which are linked to mean reversion, momentum of relative interest rates and absolute levels of interest rates. The systematic strategy focuses on the 45 currency pairs of the G10 trading only a small portion in the emerging markets to date.

The primary reason for choosing those currency weights is the lack of reliable emerging market data available for testing concepts over the last 10 years. In addition to the diversification through country focus, between the systematic and discretionary management styles, the two approaches use differing trade duration to maximise the return potential and reduce drawdowns. The models' trade durations are: 1 to 3 days; 5 to 7 days; and 7 to 175 days. This complements the trading durations on the discretionary part of the portfolio which range from less than one month for opportunistic trades, between one and three months for medium-term trades and between three and twelve months for long term strategic trades.

Whilst there are three models being used on the systematic side, Augustus are constantly reviewing the quality and consistency of these and are developing other systematic models that may be added onto and/or replace the existing models over time.

Such a combined currency product that is capable of delivering double digit returns but has a low correlation with equities is potentially a very attractive addition to an investor's portfolio especially in the current environment. On the correlation side the combined discretionary and systematic approach currently has a correlation of 6% to the S&P 500, 18% to the WGBI and an annualised return of 11.4% (net of fees). These numbers reflect three and a half years of managing client assets within other Julius Baer single manager hedge funds before the launch of JB Currency Hedge Fund in July 2007. The Fund is targeting Libor plus 7%-10% per annum.

Adrian Owens and Mark Dragten are Portfolio Managers at Augustus Asset Managers Limited