The Need for Currency Hedging in a Volatile World



The Need for Currency Hedging in a Volatile World

|[|By Anthony Golowenko, CFA, Portfolio Manager, SSgA - Australia |

|p| |

|i| |

|c| |

|]| |

| |  |

| |This essay is the first instalment of a series of three educational pieces which closely examine the topic of strategic currency |

| |management. In the current environment of volatile financial markets and more subdued expected equity returns, the issue of managing |

| |foreign currency exposure is particularly relevant when making international investments. |

| |Introduction |

| |The dramatic rise in the Australian Dollar earlier this year has led many investors to evaluate the need for, or re-evaluate their policy|

| |for, managing foreign currency exposure. Since the start of 1997 a weakening AUD has meant that Australian superannuation plans have not |

| |had to be terribly concerned with the currency risk of investing in foreign denominated assets. The second quarter of 2002, however, has |

| |served as a timely reminder that currency returns are not always a 'one-way street'. This highlights the need to review the currency, as |

| |well as bond and equity, benchmarks on a regular basis. The fact that the Australian dollar / US dollar exchange rate has not been |

| |'one-way traffic' may be seen in the chart below. |

| |Figure I: The AUD Has Not Always Been 'One-Way Traffic' |

| |[pic] |

| |Source: FactSet and SSgA |

| |Risk Equals Return? |

| |One of the fundamental concepts of Modern Portfolio Theory is that in order to achieve a higher level of return, investors must assume a |

| |greater level of risk. Via this risk versus return trade-off, investors are conditioned to believe that in return for taking on |

| |additional risk, they will be compensated with higher returns1. But, is this risk/return 'conditioning' consistent when it comes to |

| |foreign currency exposure? |

| |The simple answer is "no". Holding a foreign currency over a long time-horizon provides an expected return of zero. Considering currency |

| |in a traditional valuation framework, it provides no future cash flows in the form of dividends or coupon payments and does not generate |

| |any underlying earnings to provide capital appreciation. Currency is typically a by-product of some other transaction in equity or fixed |

| |income securities. |

| |Zero Long-Run Expected Returns |

| |The most established and well-known theories relating to foreign currency exposure providing no long-term returns are Purchasing Power |

| |Parity (PPP) and Uncovered Interest Rate Parity (UIP). |

| |Returning briefly to Economics 101, PPP has its grounding in the Law of One Price, which says that identical goods sold in different |

| |countries must sell at the same price when expressed in the same base currency. In the absence of perfect competition, trade barriers, |

| |transportation costs and other frictions, arbitrage will ensure that prices are 'bid up' or 'sold down' to the point where the forces of |

| |demand and supply are in balance. PPP extends the Law of One Price by aggregating the prices of all single goods to form the overall |

| |price level of an entire economy. |

| |UIP is based on the notion that the interest rate differential between the home and foreign markets will determine the change in the |

| |exchange rate. In very broad terms, if interest rates in Australia are 10% p.a. and interest rates in the U.S. are 5% p.a., then after |

| |one year (according to UIP) the AUD should depreciate relative to the USD by around 5% in order for the returns to be equal when |

| |expressed in the same base currency. |

| |Academic studies of these parity relationships have been mixed. They have generally found that for longer time-periods (greater than |

| |three to five years), currencies will revert to their theoretical levels. Over shorter time-periods, however, researchers have found that|

| |exchange rates depart from their theoretical 'fair value' for considerable lengths of time and their returns can exhibit non-random |

| |trends and reversals. |

| |The results for both PPP and UIP are consistent with currency exposures having zero long-term expected returns. But, as will be discussed|

| |in greater depth later, in this essay, short-term returns are a different matter. |

| |The Volatility Drag |

| |As foreign currency returns over the 'long run' have an expected return of zero, it logically follows that over time their affect will |

| |wash out, right? The answer is "not exactly". |

| |The subtlety of this concept is best illustrated by an example. Consider two hypothetical portfolios that have the returns described in |

| |Figure II. |

| |[pic] |

| |In Figure II, it may be seen that at the end of two years both portfolios have returned 10%. Well, "yes and no". From an additive |

| |(arithmetic) perspective, both portfolios have provided the same level of return, 10%. From a compound (geometric) perspective, however, |

| |the returns are quite different. Portfolio A has a compound return of 8%, while Portfolio B has a significantly higher (compound) return |

| |of 10.16%. To see how this works, let's take a look at Figure III, which shows the value of the hypothetical investment after years one |

| |and two. |

| |[pic] |

| |At the end of Year 1 it may be seen that Portfolio A has a significantly larger cash value ($120) than Portfolio B ($108). For Portfolio |

| |A, this translates to a greater monetary value that participates in the negative return that is experienced in the second year (-10%). |

| |The greater the volatility of the underlying return series, the larger the difference between the arithmetic and geometric returns will |

| |be. More specifically, the geometric annual return (Rg) is approximately equal to the arithmetic annual return (Ra) less the return |

| |series volatility (σ 2/2), [Rg ≈ Ra - σ 2/2] 2. |

| |From a passive currency exposure perspective, the implication for international investing is that for two portfolios with the same |

| |average (additive) returns, the one with the lower volatility will produce the greater compound return. |

| |In their paper of 2000, research conducted by Gorman & Qian (from a U.S. perspective) found that a 50% hedged international stock or bond|

| |portfolio can actually underperform its unhedged counterpart by 0.50% p.a. and still provide an equivalent compounded return. |

| |The Effect of Currency |

| |Figure IV illustrates the significant influence of Australian Dollar movements on the returns of the MSCI World Ex-Australia Index over |

| |the period 1981-2002 (September 30th, 2002). As described previously, economic theory dictates that the impact of currency returns should|

| |theoretically 'wash-out' over the long-term. As the Figure below illustrates, the amount of time required for short-term currency trends |

| |to reverse and offset previous foreign exchange returns may test the patience of even the longest-term investor! |

| |Figure IV: Affect of Currency on International Equity Investing |

| |[pic] |

| |Source: MSCI and Fact set where net dividends have been re-invested. |

| |The final 'bar' in Figure IV describes the average, after allowing positive and negative returns to offset one another, of the local |

| |equity and currency returns. The average currency return of 5% p.a. represents in the order of 47% of the average local equity return |

| |(10.4% p.a.) or 32% of the average total MSCI World Ex-Australia (Unhedged) Index return of 15.4%. Presented in this context, one can |

| |certainly appreciate the importance of the currency hedging decision. |

| |Not a 'Zero-Sum' Game in the Short-Run |

| |In a previous section of this essay, the economic theories describing the zero long-run expected return of holding foreign currency were |

| |presented. However unlike the 'long-run', PPP and UIP do not typically hold over shorter time-periods. |

| |Figure V presents the difference between the AUD hedged and AUD unhedged returns of the MSCI World Equity Index. Results are presented on|

| |a rolling 5-year annualised basis over the period January 1988 to September 2002. It may be seen from the Figure that over what some |

| |might refer to as the 'medium-term', there is less than a '50/50 chance' (44.1% to be exact) of the currency impact falling close to zero|

| |(± 1.0% p.a.) for the sample period. Or, to put it a slightly different way, over the sample period there is less than a 50% chance of |

| |the affect of AUD appreciations and depreciations 'washing out' over a 5-year period. |

| |Figure V: Currency Impact on Global Equity Returns |

| |[pic] |

| |Source: MSCI and FactSet |

| |Another striking feature of Figure V is the dispersion of foreign currency returns that have been observed over the sample period. This |

| |highlights the significant volatility that is present in foreign currency exposures and reinforces the currency Volatility Drag effect |

| |described earlier. |

| |Equity and Fixed Income Benchmark Conventions |

| |The issues presented in this essay thus far have focused on currency volatility not being compensated with return, the theory of foreign |

| |exchange parity relationships, the impact of volatility on compound returns and realised returns of the recent past; but what about the |

| |'real-world'? How do these topics relate to what actually happens in the investment management industry? |

| |Let's take a look at the market convention for money that is managed against international equity and fixed income benchmarks. The large |

| |proportion of international equity portfolios are managed against unhedged benchmarks. Yet for international fixed income portfolios, the|

| |majority are managed relative to fully hedged indices. Why the difference in market convention? |

| |The answer lies in relative currency volatility3. For many years global bond managers have recognised the potential for currency returns |

| |to 'swamp' the underlying asset returns as a result of currency return volatility's tendency to exceed bond return volatility. |

| |International Equity managers, on the other hand, have collectively been of the opinion that currency volatility is of much less |

| |significance relative to the underlying asset's return volatility. |

| |In the current environment of more subdued international equity returns, however, the impact of currency volatility becomes more |

| |significant. Returning to the concept of the Volatility Drag presented earlier, the currency Volatility Drag is a much higher proportion |

| |of the lower international equity returns experienced whilst in a more subdued economic environment. It is for this reason that investors|

| |need to consider implementing an international equity currency hedging program. |

| |Conclusion |

| |As we have seen recently, the fortune of AUD is not a 'one-way street'; what goes down must also go up. This essay has shown that: |

| |Additional risk is not always compensated with extra return. |

| |Heightened variability of returns increases the "Currency Volatility Drag". |

| |The impact of currency over shorter time-periods is often significant. |

| |Global bond managers typically insulate their underlying asset's returns. |

| |In the current environment of more subdued expected international equity returns, more so than in the past, serious attention needs to be|

| |given to a fund's currency hedging policy. The intention of this article is to highlight the reasons why investors must consider hedging |

| |at least part of their portfolio's international equity exposure and must maintain a long-term perspective when formulating their |

| |investment policy. |

| |Next Instalment |

| |Part II of this series of essays will examine the issue of the 'optimal' hedge ratio, so please 'stay tuned'. |

| |1 From an international equities perspective, over the period January 1989 to September 2002, the annualised volatility of the (unhedged)|

| |MSCI World Index in AUD of 16.93% is in excess of the MSCI World Index Hedged into AUD's annualised volatility of 14.31%. |

| |2 Gorman and Qian (2000), "International Benchmarks: In Support of a 50% Hedge Ratio", The Journal of Investing, Vol. 9, No. 2. |

| |3 Academic research has shown that since free-floating exchange rates began in 1973, (from a standard deviation perspective) historically|

| |30% of international equity volatility is attributable to currency while for international bonds this figure is 60%. |

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download