Currency hedging 2.0
Ugo Lancioni, Fredrik Repton and Nikola Petrovic argue that it is possible to make currency hedging more forward looking and better integrated into asset allocation
“When the facts change, I change my mind,” John Maynard Keynes is famously said to have said to one of his critics. “What do you do, sir?”
Managers of currency hedging programmes might well ask this question. Currency hedging has not changed for years, and the traditional approaches fail to take account of changing market dynamics satisfactorily.
When so much effort is expended to make asset allocation more dynamic it is odd that currency risk gets left out. This article describes our concept of the dynamic ideal hedge ratio, which aims to take currency hedging a step forward by making it forward-looking as well as fully-integrated into the whole-portfolio allocation process.
Until now, most investors have taken one of five traditional approaches to currency risks:
• Remaining unhedged;
• Hedging 100% of their exposure ‘passively’;
• Part-hedging their exposure (50%, for example);
• Employing a ‘static’ strategic hedge ratio determined by the historical relationship between the underlying portfolio and its foreign currency exposures (a refined version of the part-hedged solution);
• Hedging statically with an additional ‘active’ strategy to exploit pure alpha-generating opportunities.
All five models have drawbacks. Staying unhedged involves unrewarded risk. A full hedge can be expensive. A ‘static’ strategic hedge ratio integrates foreign-currency risks into the whole-portfolio construction process but remains a ‘set-and-forget’ approach based on historical correlations. An ‘active’ hedge adds a dynamic element, but one that is unrelated to the portfolio construction process and can create big exposures to manager risk and style risk — the carry strategy damaged many hedging programmes during 2007-08, for example.
We suggest combining both worlds: ‘dynamic’ in adapting to changing markets and ‘ideal’ because it integrates currency risk management into whole-portfolio risk management. The goal is to set a hedge ratio that is rewarding for the overall portfolio at all times.
To integrate currency risk into whole-portfolio risk management we use the Black-Litterman framework to optimise exposure to the expected returns of three theoretical ‘securities’:
• The foreign currencies that come with holding the assets (against the investor’s base currency).
• The aggregate yield differential between the base currency and the foreign currencies (that is the expected cost, or negative returns, of hedging).
• The underlying assets priced in local currencies.
By changing the combination of these securities one can derive the optimal, or ideal hedge ratio.
Making the process dynamic has two stages. First, input forward-looking expected returns into the optimisation: for the foreign currencies (1) use a purchasing power parity-based valuation approach, taking into account the extent to which currencies are mis-valued but also how long they have been misvalued; for expected hedging costs; (2) consider the short end of respective currencies’ yield curves; and for the underlying assets; (3) use the client’s own expectations or neutral indicators such as bond yields or economic growth expectations.
Second, while generating forward-looking expectations for correlations between these three is not practical, one can generate a sensitive approach by using a shorter observation window on correlations and by repeating the optimisation process regularly and frequently – here monthly optimisation is assumed. By contrast the ‘set-and-forget’ static hedge ratio can go years without being updated.
Bring all of this together and extreme foreign currency overvaluation, cheap expected hedging costs and high correlation between foreign currencies and underlying assets should generate a higher hedge ratio – and vice versa.
One can see that the hedge ratios experience substantial changes that are completely different for each base currency. There is no debate that this is ‘dynamic’ and base currency-specific – but was it ‘ideal’? Let’s look at 13 years of performance for these two investors, as well as for yen, sterling, Australian dollar and Canadian dollar-based investors, and compare that with the results they would have achieved had they invested in the MSCI World index unhedged, fully-hedged or 50% hedged (see table).
Three things stand out. The ‘dynamic ideal hedge’ ratio portfolio enjoyed a higher annualised return than the other three, with two exceptions: the JPY-based investor exposed to the MSCI World unhedged and 50% hedged. However, simple excess return is not the objective: currency hedging is about improving the risk-adjusted return of the portfolio, and measuring with both volatility and downside volatility the ‘dynamic ideal’ hedge ratio portfolio outperformed the other three for all six base currencies.
That is significant, particularly when considering why the JPY-based investor got better returns by remaining unhedged. Because the yen acts as a safe haven – strengthening during global risk aversion – it pays not to hedge back to this currency when global equities are rising.
Think how different things are for an investor based in Australian dollars, a pro-cyclical currency which strengthens when risk appetite is high. The fully-hedged Australian dollar investor far outperformed the unhedged investor.
The fact that the dynamic ideal hedge ratio improved risk-adjusted return for both these investors shows it is not dependent on the base currency’s profile. But the regular and frequent adjustment of the ratio also ensures that efficacy is not dependent on the persistence of long-term currency dynamics. Should the Japanese yen become a pro-cyclical currency tomorrow, the process will begin adjusting to the new reality within a month.
For those who deploy a traditional method of currency hedging, the answer may be to stick with an out-of-date static hedge ratio, or to trust an active currency manager to add alpha from changing market dynamics, with no consideration of the underlying portfolio risks. It is time to take the next step with a hedging process that is both dynamic and fully-integrated into portfolio risk management.
Ugo Lancioni is senior portfolio manager and head of global currency, Fredrik Repton is portfolio manager and Nikola Petrovic is associate portfolio manager, investment grade fixed income, at Neuberger Berman