When deciding on strategic asset allocations many investors grapple with the dilemma of whether or not to hedge currency risk. Historically, a typical approach was to hedge currency exposures in foreign bond portfolios, while leaving foreign equity portfolios largely unhedged. However, the lower yield environment in the eurozone relative to other developed bond markets, such as the U.S., encourages many investors both to seek higher yields abroad and to question whether hedging foreign bond allocations is still worth it.
The answer to this question clearly depends on investors’ individual risk-return preferences, with more risk-averse market participants tending to hedge most of the currency risk of bonds. However, an informed investment decision should also take into account several other factors, such as the level of currency hedging costs, expectations for foreign exchange (FX) movements and the characteristics of the underlying bond portfolios. Importantly, whether or not to hedge currency risk does not need to be a binary decision, and in some cases partial hedging can be appropriate.
Cost of currency hedging
For euro-based investors wishing to hedge U.S. dollar assets, the return drag from hedging has risen from 0% at the beginning of 2014, to over 2% today. Why has this happened?
Hedging costs can be decomposed into two parts:
- Short-term rate differential: This is the difference between short-term rates in the domestic and foreign currency. When investors wish to hedge a foreign currency exposure, they implicitly pay the foreign cash rate and receive the domestic cash rate (known as covered interest rate parity). So any changes between these rates impact the hedging cost.
- FX/currency basis: This is the additional cost investors pay to buy and sell currencies forward, on top of the rate differential. This factor largely depends on imbalances between supply and demand across different currencies.
As Figure 1 shows, both of these components have expanded relative to four years ago. The differential between U.S. dollar and euro short term rates (in grey) has grown as a result of diverging Federal Reserve and ECB policy rates. At the same time, the FX basis (in blue) has remained negative due to strong net demand for U.S. dollars globally.
However, as the chart also shows, the current level of hedging cost is sizable but not unprecedented. Investors faced comparable if not higher hedging costs in 1999-2000 and 2005-2006, periods in which the Federal Reserve also raised interest rates (and also very briefly when Lehman Brothers defaulted in 2008). One important difference, however, is the starting level of yields: 2% hedging drag when 10-year U.S. Treasuries yield close to 2.5% is far harder to cope with than when they yielded more than 4%, which was the case in previous periods of high hedging costs.
Time is unlikely to solve the dilemma and, on the contrary, may make it worse. If the Federal Reserve continues its hiking cycle ahead of the ECB, the differential between U.S. dollar and euro short-term rates could widen further.
Benefits of hedging high quality bond portfolios
Does this mean investors should stop hedging currency risk in their bond portfolios? We do not think so, and to see why, it is important to consider the impact of hedging on risk as well as return. Comparing unhedged and hedged U.S. Treasury investments since the launch of the euro (shown in Figure 2 and 3) highlights two key points:
- Leaving high-quality bond allocations unhedged (blue line) materially increases risk relative to a currency hedged investment (yellow line). Although a euro-based investor would have improved cumulative returns by not hedging currency exposure, the much higher volatility – driven by currency risk – would have halved risk-adjusted returns, represented by the Sharpe ratio.
- Although saving 2% per year in hedging costs may sound appealing, these savings can be easily wiped out by currency volatility. In fact, historically the U.S. dollar has depreciated more than 2% versus the euro almost half of the time on a one-year rolling horizon (green bars).
Currency hedging and bond characteristics
The removal of currency hedging can profoundly change the risk-return profile of bond allocations. But does this equally apply to all types of bonds?
It does not, because the additional risk generated by foreign currency exposure depends on two different factors:
- The correlation between the portfolio and foreign currencies. High quality currencies such as the U.S. dollar often tend to strengthen when riskier bonds like high yield are under pressure. So for euro-based investors in U.S. high yield, not hedging U.S. dollar exposure may reduce portfolio losses in periods of market stress, as the currency gains may partially offset the underlying high yield losses. In general, the lower the correlation between underlying asset and foreign currency, the lower the incremental risk generated by leaving currency exposure unhedged, since a larger share of the risk is diversified away (see the illustrative example in Figure 4).
- The risk of the underlying portfolio. The higher the volatility of the underlying bonds, the lower the incremental risk introduced by leaving currencies unhedged. As an illustrative example (see Figure 4), consider a currency with 10% volatility, uncorrelated with the underlying bond portfolio. If the volatility of the hedged bond portfolio was 2%, unhedged currency risk would increase it by 8.2%. If the bond volatility was 10%, currency risk would increase it by only 4.1%.
Partially hedged portfolios
Similar considerations also apply to partially hedged portfolios. Figure 5 shows how the long-term expected return and estimated volatility of a U.S. Treasury and U.S. high yield portfolio vary depending on the hedge ratio, from the perspective of a euro-based investor1.
For both asset classes removing currency hedging is expected to increase returns, as savings of direct hedging costs are not expected to be eroded by fluctuations in the value of the U.S. dollar over the long term. However, consistent with the above, the impact on volatility is much larger for Treasuries than for high yield bonds. Partially hedging low-quality bond portfolios, such as high yield, may help enhance returns without necessarily adding – or even slightly reducing – volatility relative to a fully hedged allocation. This is because high-quality currency exposure may help diversify the underlying credit risk.
A final point is that the above results can differ if currency hedging is conducted at the total portfolio, or multi-asset level, rather than the individual asset class level. Multi-asset hedging goes beyond the scope of this paper but further details can be found in “Currency Hedging Optimization for Multi-asset Portfolios” by Helen Guo and Laura Ryan, and “Optimal Currency Hedging: A Factor Perspective” by Steve Sapra and Lutz Schloegl. Ultimately the chosen currency hedging framework should take into account investors’ specific portfolios and objectives.
In short, when it comes to currency hedging decisions for bond portfolios, investors should carefully consider multiple dimensions, including not only hedging costs and expected FX movements, but also the specific credit risk of the underlying assets, the related currency exposures and their mutual interactions. They should also ensure their hedging policies are consistent with their specific return targets and risk budgets. Even in the presence of high hedging costs, currency hedging bond portfolios is often still appropriate, although in some cases partial hedging may be beneficial.