When overlays add value
Adding a currency overlay program to a portfolio with cross-border investments is the preferred way to manage currency exposures. Pension consultants, who have examined currency overlay managers and their returns, support and recommend that investors initiate currency overlay programs to reduce currency losses and to increase returns.
In May 2000, Frank Russell Company, a consultant, published a study of currency overlay returns. It found that the average excess return of the 241 accounts was 1.06% per year from December 1988 through June 1999. In a performance update, Frank Russell has reported that the average, annual added value in its “all currency, all hedge” universe of 19 overlay managers was 1.22% for the six years through 2002. Currency overlay managers have continued to add value.
Although currency overlay has proved its mettle, it is sometimes seen as “unnecessary” or “too complicated” even though the added value it can provide is worth the effort (see IPE September 2000). In an age of diminished equity returns and historically low fixed income yields, the potential to reap excess returns that have averaged over 1% per year since 1988 cannot be ignored.
Surprisingly, even though consultants are aware of the large impact currencies have on portfolio returns, they continue to search for international equity managers with little thought of how the currency risk should be managed. It is rare that a search for an international mandate is a combined search for an equity manager and a currency manager.
In a review of over 100 EAFE managers by Deutsche Bank’s InterSec Research Corporation it emerged that “no equity manager added value through active currency management in the past 10 years”. They simply lack the skills to manage currencies effectively. And, many of them do not hedge at all. That is why searches to fill international mandates must be expanded to include a currency overlay manager.
In addition to allowing equity managers to hedge currency exposures or to leave them hedged or unhedged at all times there are other mistakes that need to be avoided.
Most pension plans have strategic allocations to each asset class in their portfolios. Consultants therefore recommend that investors initiating a currency overlay programme begin by setting a strategic currency allocation. The strategic hedge ratio should be determined by relating the return and risk characteristics of currencies to those of the other assets in a portfolio. Sounds simple, but it is not.
A large body of research has demonstrated that there is no universal hedge ratio that provides an optimal risk/return profile. The optimal hedge ratio is therefore considered as specific to each investor’s asset mix and risk tolerance.
To set the strategic hedge ratio, consultants and plan sponsor can feed their optimisers with a portfolio’s asset mix and assumptions about currencies. It is usually assumed that:
1) currency has no expected return,
2) the currency return is independent of the underlying asset return, and
3) there is no impact on a portfolio from cash flows associated with passively hedging all or a portion of the currency exposure.
Those assumptions are incorrect. However, since many pension plans with passive or active overlay programmes have made them, they will discover that the results will deviate significantly from those indicated by their optimisations.
Currencies are assumed to have no expected return because they do not pay dividends, interest or have a claim on assets that may appreciate over time. However, it is an irrefutable fact that currencies have short- and long-term returns that can be large.
The dollar’s value versus the German mark, from 1970 to 1998, and then against the euro, for example, has risen or declined by 10%, or more, in 60% of the past 30 years. It is this random volatility that investors seek to reduce by managing their currency exposure.
Long-term currency returns should not be ignored when the strategic hedge ratio is determined. As tabulated in ‘Triumph of the optimists – 101 years of global investment returns,” a stdy published by Princeton University Press, the Deutsche-mark had an annualised appreciation of +2.1% against the dollar from 1950 to 2000. The Japanese yen rose at an annualised rate of +3.3%, while the pound fell at a rate of –0.9% over the same fifty years (see table).
Most international equity investments in US pension plans are managed against MSCI’s EAFE index. The euro, sterling, and the yen dominate EAFE’s currency exposure, adding up to 82% of the total exposure. The currency return of the EAFE index has thus been positive over the past 30 years. It is therefore incorrect to ignore history and assume a currency return of 0% when the strategic currency exposure is determined. Using a currency return of 2% is more realistic and will result in a more accurate hedge ratio for dollar-based investors.
An annualised long-term currency return of 2% for US pension plans investing in EAFE is not trivial when cash and fixed income yields are at their lowest in a generation. And, long-term currency returns are not the same for investors with the euro or the yen as the base currency. Japanese investors have seen the dollar drop about 3% per year, on average, since 1950.
The currency returns of the past 50 years suggest that dollar-based investors need to have a strategic currency benchmark that is largely unhedged, while European and Japanese investors need to have currency benchmarks that are largely hedged.
However, since it has been impossible to find a universal hedge ratio many dollar-based pension funds use a 50% hedged currency benchmark. Currency managers and consultants tend to favour the 50% hedged benchmark as well. Not because it is optimal, but because it is easier to add value against it.
Will the historical currency returns persist? We believe they will since numerous academic studies have concluded that exchange rates are largely determined by relative inflation rates over the long-term - the Purchasing Power Parity theory (table). It seems that it is institutional differences between nations that determine if a country will tolerate more or less inflation. Since political institutions do not change much, it is likely that the US and the UK will continue to have higher average inflation rates than Euroland and Japan. It is therefore reasonable to expect that the dollar will continue to depreciate over the long-term.
Currency returns are not independent of underlying asset returns. This complicates the process of finding a strategic hedge ratio. Because consultants and investors usually assume that currency returns are 0%, they believe they can ignore the fact that realised currency returns in investment portfolios depend on the underlying asset returns. The optimal hedge ratios they calculate will therefore be way off the mark. An example for a dollar-based investor will illustrate why.
Assume that the annualised long-term EAFE equity return is 8% in the decade ahead. At the end of 2013, $10m invested in EAFE will have grown to $21.6m for a cumulative equity return of 116% assuming the currency return was 0%. However, if the value of the underlying currencies appreciates at an annualised 2%, then the value of the investment will top $26.3m in 2013. That is an additional gain of $4.7m or a cumulative currency gain of 47%.
Measured independently of any underlying asset, a currency return compounding at 2% will become +21.9% in 10 years. However, since that currency appreciation impacts an initial investment and its increased value, assumed to grow 8% annually in our example, the cumulative portion of a portfolio’s total return that is caused by the currency appreciation is +47.2%. That is a currency gain of 4% per year when annualised over 10 years.
The example illustrates that assuming a 0% currency return that is also independent of the underlying asset return, is a vast simplification in the optimisation process to set the strategic currency exposure. It will lead to a hedge ratio that is far from optimal.
Setting the hedge ratio, regardless of the assumptions made, usually results in an outcome that indicates a certain percentage of a portfolio’s currency exposure should be hedged at all times. It is assumed that passively hedging that portion will not meaningfully impact the underlying portfolio. Thus, the cash flows associated with a passive hedge are simply ignored. That is an error that has been costly for several euro-based investors with large dollar exposures.
Several pension funds in Euroland initiated passive hedging programmes in the 1990s that covered half or all of their currency exposure. When the dollar rose from 1995 into 2000, the hedging losses on a typical exposure of 60% dollars and 20% each in yen and sterling, became very large. In fact, the value of the underlying currencies rose 55% while the cumulative hedging ‘cost’ of a negative interest rate differential was 10%. If the exposure was hedged 100%, the cumulative hedging loss became 65% of the underlying portfolio’s initial value (see chart).
Although the passive hedging loss was offset by the underlying currency appreciation, whenever hedges were renewed, the hedging losses of the preceding period had to be paid. Cash could have been used at first, although some funds may have borrowed to
settle the losses. However, at some point it would have become necessary to sell some of the underlying assets to raise cash to pay the hedging losses. Thus, these passive hedges impacted the underlying assets by incurring interest rate charges, and/or the holdings of the underlying assets were involuntarily reduced when they were sold to settle hedging losses.
The dollar’s rise is now permitting these investors to recover some of their passive hedging losses. However, faced with the problem to determine a strategic hedge ratio and passively hedging to that ratio, investors face the question; is there a better alternative?
The alternative is an active currency overlay programme in which each currency is hedged when it declines and is unhedged when it appreciates. As several studies of overlay returns have demonstrated, overlay managers add value to their benchmarks. It means that active overlay managers are largely correctly hedged when currencies decline with hedges that are profitable (see chart). It also means that they are largely unhedged when currencies rise, permitting their clients to reap those gains.
Investors who have passive hedging programmes can avoid large cash losses when currencies rise and improve their returns by switching to an active overlay programme. Investors with no overlay can boost their returns by adding an active overlay programme.
Simply put, active currency overlay managers can produce superior return and cash flow profiles compared to the passive alternatives of never hedging or always hedging a portfolio’s currency exposure.
Alfred Bisset is president and Ulf Lindahl is chief investment officer of AG Bissett & Co in Rowayton CT