Most Popular Commentaries

Executive Summary

Part one of this series (Currency Hedging: Why, When and How — December 2009) focused on the factors behind the decision of whether or not to hedge foreign currency exposure and is a useful introduction to many of the basic concepts discussed here.

In this second installment of our Currency Management Series, we make a case for the value of active currency management. We believe there are opportunities for professional currency managers to extract alpha on a consistent basis.

Click here to view Part One of the Currency Management Series: Currency Hedging

We begin with a brief overview and history of the currency market. Second, we illustrate why the currency market, although the deepest and most liquid market in the world, is inefficient due to a majority of participants being passive investors seeking liquidity and currency hedging to reduce risk exposure. Next, we review historical studies and results, which confirm that active currency management has been a fairly consistent source of excess return over time for skilled, profit-oriented investors. Finally, we show that currency as an asset class and source of alpha is very attractive in a traditional asset allocation framework. Not only have currencies exhibited low correlations to other asset classes, but also these correlations have remained stable even during periods of financial crisis or turmoil.

Click here or call 800-345-6488 for more information on products and services available to you

Since the shift to the flexible exchange rate system in the early 1970s, the currency market has exploded in terms of growth. The primary reason is increasing globalization of the economy that has led to rapid growth in cross-border trading. As Figure 1 illustrates, growth in international trade as a percent of global GDP has mirrored the increasing level of GDP itself. By 2007, nearly one-third of global GDP was generated through international trade. To contrast relative market size by average daily turnover, the global currency market, at $3.1 trillion, dwarfs the daily turnover in the major bond markets ($905 billion) and stock markets (approximately $322 billion) around the globe.1 Given the sheer size and depth of the currency market, the reader might be tempted to conclude it is very efficient in terms of pricing.

Is the Currency Market Efficient?

In what many consider to be an essential reference book, The Economics of Exchange Rates 2 authors Lucio Sarno and Mark Taylor summarize the overwhelming consensus, based on thirty years of research, that the currency market is not efficient by the classical definition. The consensus rationale is that the Efficient Market Hypothesis (EMH) relies upon several strong assumptions, none of which is especially valid in the currency market. These assumptions include that market participants are 1) homogeneous (i.e. they all have the same goals and objective in the market), 2) rational (i.e. they all have the same model for how the markets works and make use of all publicly available information) and 3) risk-neutral (i.e. they care only about expected return and not risk).

In an academic study titled Under the Microscope: The Structure of the Foreign Exchange Markets ,3 authors Michael Sanger and Mark Taylor argue that market participants are far from homogeneous, rational and risk neutral. On the contrary, they state that the currency market incorporates a heterogeneous set of participants with differences caused by informational asymmetries, different reaction speeds to significant data innovations, diverse opportunity sets and risk-return expectations. To illustrate some of these heterogeneities, the authors characterize participants as either active or passive in reference to the goal they are trying to achieve when trading in the currency market.

Passive participants are exposed to foreign exchange as a result of the sale or purchase of underlying assets such as international equities or bonds or from the accrual of international revenues and costs by corporations. They do not seek to maximize the foreign exchange profit when they trade in the currency market. On the other hand, active participants participate in the currency market with the goal to maximize their foreign exchange return. Examples include currency overlay firms, Commodity Trading Advisors (CTAs) and hedge funds.

The existence of numerous and sizable participants who do not try to maximize their foreign exchange profit is a differentiating factor of the currency market compared to other financial markets. Large corporations doing business beyond their national borders that need to convert revenues and earnings back into their home currency for financial reporting and cash flow purposes is one example of passive participants trading very large volumes in the currency market. Central banks and national governments whose objectives are to implement the monetary policy of their respective countries are another example. Deutsche Bank, a leader in foreign exchange trading, estimates that passive participants account for at least 50% to 75% of the currency flows. In other words, profit maximization participants are a minority in the currency market.4

Sanger and Taylor conclude that the behavior and interaction of these heterogeneous groups of participants generates persistent inefficiencies and ensures that the process of price discovery is opaque. These inefficiencies create opportunities which professional investors can exploit on a consistent basis, as demonstrated by the facts presented in the following section.

Currencies as a Source of Alpha

There is abundant evidence that supports the case that professional investors successfully exploit the inefficiencies in the currency market.

A report by the Reserve Bank of Australia, in the paper, The Profitability of Speculators in Currency Futures Markets ,5 concluded that speculators (i.e. those engaged in the currency market to earn a profit such as professional active currency managers) earned profits over the 1993 to 2003 time period. They studied the net positions of speculators (as registered with the Commodity Futures Trading Commission, or CFTC), who traded in six currency future contracts relative to the U.S. dollar (the Australian dollar, British pound, Canadian dollar, euro/German mark, Japanese yen and Swiss franc) based on trading at the Chicago Mercantile Exchange (CME). They also concluded that the level and consistency of speculator profit increased as they engaged in trading in multiple currencies as opposed to just one or two. Diversification across currencies by professional active managers enhanced the consistency of their profits over time.

Similar results were found in a Deutsche Bank research report Currency Markets: Is Money Left on the Table? .6 This report introduces a model for the behavior of participants in the currency trading markets defining two basic participants: Liquidity Seekers and Profit Seekers. Liquidity Seekers are defined as commercial enterprises whose primary function is multi-national business management. Their goal in the currency market is to hedge exposure to foreign currency fluctuations. Profit Seekers are professional active currency managers, whose sole objective in the currency market is to earn a profit based on exchange rate movements. In the words of Sanger and Taylor in the paper Under the Microscope: The Structure of Foreign Exchange Markets mentioned in the previous section, liquidity seekers would be passive investors and profit seekers would be active investors.

According to the report, Profit Seekers are able to consistently profit by providing short-term liquidity and passive hedging needs to Liquidity Seekers. The authors analyzed profits earned by both groups using currency future contracts traded on the CME from 1993 to 2006 based on data from the CFTC. It estimated that in each of the fourteen years studied, Profit Seekers consistently earned profits at the expense of Liquidity Seekers.

An analysis of the historical returns achieved by active currency managers strongly supports the conclusion that professional currency managers can exploit the inefficiencies in the marketplace as well.

Several institutions (banks, investment companies and investment consultants) have maintained indices that have tracked currency manager performance for up to 30 years. One example is the Barclays Capital Currency Traders Index, which is an equal-weighted composite of managed (i.e. active) currency trading programs employing currency futures and/or forward contracts in the inter-bank market, maintained by Barclay Trading Group Ltd. As shown in Table 1, since its inception in 1985, this index delivered a total return of 7.7% per year with a standard deviation of 11.7%. This is a total return index, net of management fees.

To have a better idea of the managers ability to add value, we need to subtract from the total return the risk-free rate and add the management fees. Adjusting for the risk-free rate (we use three-month U.S. T-bills), the index has earned a 3.3% annual excess return with 11.7% volatility, which equates to a 0.28 information ratio. Adjusting for management fees is not easy to do because they differ by manager and have changed over time. However, as a rule of thumb, by adding a 1% annual management fee the information ratio increases to 0.37. With a 2% annual management fee, the information ratio rises to 0.46.7

A second example comes from Mercer Investment Consulting, which maintains two databases on currency manager performance, one for currency funds and another one for currency overlay programs. Mercer does not publish performance indices, but in an article called What Is Active Currency Management? ,8 the company calculates a three-year rolling average information ratio (gross of fees), which ranges between zero and 1.4 overall and 0.5 to 1.8 for managers in the upper quartile of performance. In the article, the firm argues, Even the median manager has managed to add value, although the most recent period has been more difficult. There are few other active strategies that can claim that level of positive alpha. It concludes, Active currency management has proved to be a good source of added value for investors in the past. There is no reason to believe that this should change, as a material proportion of market participants are price takers and currency markets are the most liquid with the lowest dealing costs of any asset class. However, appointing a suitable active currency manager is critical to unlocking the potential gains.

There are other sources of currency manager returns, Daniel B. Stark Currency Trader Index and DB FX Select Index among them.9 Not all of these sources are publicly available, but those that are available present similar numbers that are not that different from those in the previous two paragraphs.

An additional piece of evidence supporting the existence of excess return potential in the currency market is the Deutsche Bank Currency Returns (DBCR) Index, an equal-weighted blend of the most widely used investment strategies among active currency managers (valuation, momentum and carry see text box to the right for a brief description of each of these strategies). This is an excess return index, which includes transaction costs but is gross of fees and is published daily by Deutsche Bank. The index is not composed of active managers actual excess returns, but we believe it is a good indicator of how currency managers have performed given it is a fair representation of common active strategies used by professionals in this field.10 As shown in Table 2, the DBCR Index has delivered an information ratio of 0.73 over the last 20 years with a volatility of approximately 5% and worst drawdown of 7.2%.

Given the heterogeneity of construction for these various indices (total versus excess returns, gross versus net of fees, risk-adjusted or not), the different rules for managers to be included and excluded from an index, and the typical survivorship bias of these type of active managers indices, the results are not easy to compare. Nonetheless, it is significant that several independent and well-respected sources all conclude that active currency managers as a group have added alpha over time. All these indices have posted positive information ratios on a fairly consistent basis some for as long as 30 years.

To summarize, there is broad evidence supporting the case that currencies are a source of excess returns. Based on our 20+ years of experience, we believe the type of excess results presented by all these sources are achievable in the currency market.

Other Advantages of the Asset Class

The existence of excess return potential is not the only attractive element of this alternative asset class. There are others.

• First, the currency market is extremely liquid, which means that transaction costs are very low even during times of financial crisis and turmoil when other asset class markets can demonstrate a severe loss of buyers or liquidity providers.
• Second, given the size of the daily turnover in the currency market, it is very difficult for an active currency management strategy to run into capacity constraints.
• Third, currency alpha is easily portable. Investors can add an active currency overlay to any type of portfolio; all they need is the ability to trade plain vanilla currency derivatives such as futures and/or forwards. In the third installment of this Currency Management Series, in addition to presenting our approach towards currency alpha, we will also explain in practice how an investor can overlay a currency strategy to traditional fixed income and equity portfolios.

Last but not least, currency returns have very low correlation to those of other asset classes. Therefore, we not only believe that currencies are a very attractive asset class where the manager can add alpha, but also that this alternative asset class makes perfect sense in the context of an asset allocation framework. In the next and final section we address this concept.

Currencies in an Asset Allocation Framework

Over the last three decades, numerous academic and industry papers utilizing some of the contributions by Modern Portfolio Theory have argued in favor of incorporating alternative asset classes to traditional equity and fixed income portfolios. The main rationale is straightforward: Combining assets with low correlations between each other reduces the overall risk in the portfolio for any given level of expected return. The primary alternative asset classes studied have been U.S. investment-grade credit, U.S. high-yield credit, international and emerging equity markets, international and emerging fixed income markets, private equity, hedge funds and commodities.

However, the recent global financial crisis has reminded investors of one of the limitations of Modern Portfolio Theory. Correlations among different asset classes are not stable. While a few of the aforementioned asset classes did bring some diversification benefits to traditional portfolios, during periods of market shocks and panics, their performance was much more highly correlated to equities than most had assumed or expected during normal periods.

Tables 3, 4 and 5 present cross-asset correlation matrixes over different time horizons for selected risky asset classes, including currencies.

Table 3 displays the cross-asset class correlation matrix for U.S. equities, global equities, commodities, hedge funds, emerging markets fixed income, foreign exchange, U.S. high-grade fixed income and U.S. high-yield fixed income asset classes over the January 1994 to March 2010 time frame. Table 4 displays the cross-asset class correlation matrix for the same asset classes over the June 2007 to June 2009 time frame, a period when global markets experienced severe stress due to the global financial crisis.

During the 2007-2009 period (Table 4), correlations between alternative asset classes and equities, both U.S. and global, were significantly higher than their longer term average (Table 3). For example, during the 2007-2009 financial crisis, the correlation between global equities and commodities was 53%, while that same correlation was only 29% in the period 1994 through 2010.

Table 5 calculates the differences in cross-asset correlations between the 2007-2009 time frame of financial turmoil (Table 4) and their long-term averages from 1994-2010 (Table 3). In all cases, except currencies. these alternative asset classes significantly increased their correlation to U.S. and global equities during the last financial crisis. Hence, the diversification benefits were not as much as expected because financial markets struggled through a challenging period brought on by the global crisis.

Over the full sample (1994-2010), currencies stand out as the alternative asset class with the lowest correlation to both U.S. equities (0.19) and global equities (0.15). Furthermore, it is the alternative asset class with the most stable correlation, as shown in Figure 2, which presents approximately 15 years of rolling correlations between U.S. equities and selected alternative asset classes. Its three-year rolling correlation to U.S. equities ranges from -5% to 48%, whereas ranges are wider for other alternative asset classes with relatively low long-term correlation to U.S. equities such as commodities (range is -18% to 52%), U.S. high-grade (range is -19% to 64%) or EM fixed income (range is 14% to 82%).

In sum, during the recent global financial crisis (and previous financial crises as well) currencies as an asset class did a better job than most other alternative asset classes in providing diversification benefits when combined with traditional asset classes. The correlations of currency returns to traditional asset classes returns have been both low and relatively stable, even during times of market upheaval and distress. This characteristic, in combination with the alpha potential, make currencies very attractive for enhanced portfolio diversification within an asset allocation framework.

Growing Interest in Active Currency Management

Interest in active currency mandates has increased substantially over the last decade as highlighted by the growth in assets under management by managers included in the Barclay Currency Traders Index (Figure 3).

Anecdotal evidence presented by specialized publications also point in the same direction. According to Investment & Pensions Europe (IPE). a leading European publication for institutional investors and those running pension funds, growing allocations to foreign asset classes over the past decade have led to a resurgence of interest in currencys alpha preservation capabilities, and in recent years, it has reached some prominence as an asset class in its own right. The publication argues that the potential to generate absolute returns uncorrelated to other asset classes has in part led to heightened interest and that this view has strengthened during 2008 as the equity, fixed income, and property markets became increasingly correlated. To conclude, the publication states currency markets have proved that, regardless of the global economic climate, they remain highly liquid relative to other asset classes and responsive to changes in underlying fundamentals. That right managers, utilizing a flexible approach and, most likely, focused on the analysis of fundamental drivers, will be able to continue to generate alpha in the forthcoming years and so add considerable value to a portfolio.11

As part of its 2008 Asset Management Survey, Investment & Pensions Europe reported $606 billion in currency overlay assets and $84 billion in currency managed accounts and funds as of March 2008.12

Despite the huge size of the currency market, it is not an efficient market in the classical sense. A majority of participants are passive investors whose motivation is based on factors other than profit-maximization, creating persistent inefficiencies in terms of price discovery. We have shown that historical performance supports the thesis that professional investors can take advantage of these inefficiencies and that therefore currency is an asset class where the skilled professional managers can add alpha on a consistent basis.

We have also shown that currency is a particularly attractive asset class within an asset allocation framework. The low and more stable correlation of currency to equities (compared to other alternative risk seeking asset classes) make it a very attractive addition to asset allocation models when it is implemented as a relative value strategy not subject to base currency constraints. And among alternative risk-seeking asset classes, currency stands out for its behavior during crisis.

The opinions expressed are those of John Lovito and Federico Garcia Zamora and are no guarantee of the future performance of any American Century Investments portfolio. For educational use only. This information is not intended to serve as investment advice.

You should consider the funds investment objectives, risks, charges and expenses carefully before you invest. The funds prospectus, which can be obtained by calling

1-800-345-6488 or by visiting americancentury.com/ipro, contains this and other information about the fund and should be read carefully before investing.