Beyond a ‘zero sum game’

How currencies can add alpha and lower risk

Originally published in the March 2007 issue

In the financial markets, as in many other aspects of life, misinformation can be costly. This is especially true for the currency markets, where the conventional wisdom embodies a good deal of misinformation. If this misinformation is accepted at face value it can lead investors to overlook or ignore significant opportunities for alpha generation and risk reduction.

Provided below are three elements of conventional wisdom where we believe the wisdom ends and the misinformation starts. They are:

Myth #1: Currency markets are incredibly efficient; looking for value is a fool’s errand. Myth #2: Foreign assets should be left unhedged to improve diversification. Myth #3: Currency is a zero sum game.Whatever someone earns, someone else loses. On average we all break even.

Myth No.1

Currency markets are incredibly efficient; looking for value is a fool’s errand.

This piece of conventional wisdom is often cited to support an assertion that the attractive track records of active managers can be attributed to simple timing and luck. Let’s take a look at what is efficient in the currency markets and what is not.

First, we can all agree that spot and forward pricing in the currency markets is efficient, accurately reflecting the prevailing interest rate differentials among currencies. An example illuminates this. In the spring of 2005, the UK had higher short term interest rates than the U.S. So, the logical “naive” trade would be to swap dollars for pounds and invest them for, say, three months, to capture the premium available from UK rates. The problem with the naive trade, according to economic theory, is that over those three months the dollar is predicted to strengthen against the pound by an amount that would offset the premium earned by investing in pounds instead of dollars. (Remember, by being short the dollar, the loss from the dollar’s appreciation will be realized at the end of the three months.)

But what if you are a not-so-naive investor who looks at the situation and says: “I will go long the pound and capture the short-term rate differential over the dollar. But I will also go long the dollar in the forward market to hedge my short dollar position over the three months.” You would get an A for effort but the strategy violates the “no free lunch” rule. If the dollar/pound spot rate is 1.90 (given the interest rate differential we’ve stipulated), the forward rate would be lower-say 1.895. In other words, you will get fewer dollars at the end of the contract (compared with what you could have had in the spot market) and your loss will equal the premium you gained by investing in pounds vs. dollars. This relationship is the core of what makes spot and forward pricing in the currency markets efficient.

However, it turns out that our naive investor wasn’t quite so naive. Investing in a currency with high real interest rates at the spot price, over time, has allowed the investor to earn the differential, as discussed above. Further, the high interest rate currency-the pound in our example-has tended to appreciate, not depreciate as conventional wisdom and theory predict. The reason for this inefficiency is that cash markets around the world are not integrated because central banks and global corporations have their own agendas. This is addressed further in connection with myth #3, but the discussion above should help establish two points:

  • forward markets are poor predictors of spot price direction, and
  • currency markets are efficient in equilibrating spot and forward prices but inefficient in equilibrating diverse global cash markets.

Myth No.2

Foreign assets should be left unhedged to improve diversification.

On the surface, this assertion is borne out by the comparison in the Exhibit 1 on the next page. The correlation between U.S. stocks and hedged is 0.73, while the correlation between the two with the UK is 0.67. Ergo, the latter portfolio should be less volatile according to that in a two-asset portfolio, substituting an asset with lower correlation volatility of the portfolio versus an asset with higher correlation.

The problem here is not the rule but the assets. When we substitute unhedged UK stocks for hedged UK stocks, we aren’t simply substituting one asset for another- we are adding exposure to the pound as a third risky asset. As we see in Exhibit 2, the second portfolio can be restated as a combination of the first (U.S. stocks and UK stocks hedged) plus exposure to the pound. Addition of the pound could reduce volatility, but only if it were sufficiently negatively correlated with stocks (i.e., between 0 and -1) as opposed to uncorrelated (i.e., between 0 and 1). Historically, currency tends to be uncorrelated with stocks rather than negatively correlated. (Remember hedging works because it is, by definition, introducing a negatively correlated asset, and shorting an asset has a -1 correlation.)

Looking again at Exhibit 1, the last two lines show us that over the past 16 years, U.S. stocks plus UK stocks hedged has indeed been the less volatile of the two portfolios we have discussed.

This conclusion does not imply that it is never appropriate to leave foreign assets unhedged. For example, a pension plan with inflation-sensitive liabilities may wish to leave its foreign assets unhedged, as a hedge against the weakening of the base currency due to inflation. Or, a company with foreign-denominated liabilities may find it worthwhile to leave like-denominated assets unhedged.

Myth No.3

Currency is a zero sum game. Whatever someone earns, someone else loses. On average we all break even.

If you care only about alpha then #3 is true-for every manager who wins, another one loses. But this one-dimensional statement ignores risk, which is always relevant in financial markets, and especially so for currency, where many participants are in it only for risk reduction.

Once we factor in risk, we can establish in a straightforward manner that it is a positive-sum game-meaning that two parties can trade with each other and both end up better off. Going back to the U.S./UK comparison in Exhibit 3, each investor has exposure in assets of the other country. By swapping those exposures, both U.S. and UK investors have reduced their risk. Once you have a positive-sum game, the game can take different forms, with some people willing to carry that risk in exchange for greater potential return.

Besides players motivated by risk reduction, there are other participants in the currency markets who are not incentivized by profit maximization, such as central banks. Central banks are key in setting cash rates. When you think about it, currency markets are simply cash markets, and no matter how big the positions are that we and our competitors push around, it’s not enough to affect short-term rates in the U.S. And, obviously, the U.S. Federal Reserve is guided by its agenda of noninflationary growth. If its objective were maximizing profits, then it would set rates high enough to earn a spread off of the non-interest bearing reserves it collects from member banks. Indeed, in Australia, where rates are considerably higher because of inflationary concerns, the central bank had record profits last year.

Large corporate hedgers are another category of players who typically aren’t in thecurrency markets to maximize profits. Consider a multinational car maker with contracts and obligations in foreign currencies. If the company can lock in a certain exchange rate they insulate their potential profits from the whims of the currency market. In effect, it is using the forward market to buy insurance, and they may be very willing to pay up for it. On the other hand, one of the interesting quirks of the marketplace is that sometimes hedgers can even get paid to reduce their risk.

A good example is today’s U.S. investor with a portfolio of euro-denominated assets. As of May 2005, three month U.S. dollar LIBOR yielded 3.21%, compared to 2.13% for the three month euro LIBOR. By hedging back into dollars through the forward market-and selling the lower-yielding euro-that investor reaps the 1.08% differential as a premium, as described in #1. On the other hand, in January 2003, when the dollar yielded less than euro rates, the same hedging strategy would have cost the U.S. investor 1.48%.


So we have a market with central banks, corporations and other risk minimizers who have objectives other than profitable trades. That leaves the door open for counterparties-foreign currency desks, hedge funds, and active currency managers-who are willing to assume that risk and get paid for it in their search for alpha.

How do we determine expected currency appreciation? We start with our recurring theme of interest rate differentials. As you might expect, countries with higher real interest rates attract the capital flows, which tends to strengthen those currencies. Central banks might be expected to counter this imbalance, but as we have seen, their monetary policies often drives them to do otherwise.

As investors seeking alpha in the currency markets we have to ask: what is the natural brake on those strengthening currencies, and when will it occur? Exchange rates are driven by supply and demand, which are expressed through three kinds of flows:

  • Capital flows (fast game)
  • Goods and services flows (slow game)
  • Speculative flows

The first is one we have explored-capital flowing to higher yielding currencies, in a positive feedback loop that drives appreciation. Goods and services are a different story since people will not turn on a dime and switch their foreign sourcing because of a small change in exchange rates. For a U.S. importer who gets supplies from Birmingham, UK, if the pound gets 5% more expensive, it is not likely to change the relationship because the increase may not persist. But as the pound strengthens by more than 25% over a couple of years as it has recently, a supplier in Birmingham, Alabama is going to start looking increasingly attractive.

The bottom line is that foreign currency trends can persist long enough to represent a significant source of alpha. Lets examine the situation in January 2003, when all the ducks lined up (Exhibit 4). Three currencies-the Canadian dollar, the Australian dollar and the British pound-all had significant yield advantages over the U.S. dollar. Our goods and services flow signals-which represents the percent expected appreciation of the currency-all were strong. And, on the bottom line, one year later, the appreciation of all three exceeded our estimates.


To sum up, the inefficiencies and imbalances that present themselves in the currency markets warrant the use ofan active manager. While the universe of overlay managers is really not that old, over the past five years, the median manager has added about 91 basis points.

Our discussion of hedged versus unhedged portfolios suggests that the baseline exposure should be 100% hedged, to minimize volatility, while using an active overlay manager to capitalize on opportunities that present themselves. Far from being a zero sum game, currencies represent a positive source of potential alpha and risk reduction for institutional investors.