Since the collapse of the Bretton
Woods system of fixed exchange rates in the 1970s, professional investors have
made enormous sums by speculating on currency moves. But financial advisors
never viewed foreign exchange trading strategies as an asset class, and
certainly not one that investors should allocate large amounts of capital to.
But this is changing as a growing body of research supports the idea that forex
strategies can be beneficial.
Some firms go as far as to suggest that investors put a
considerable percentage of assets into currency-based strategies. Bilal Hafeez,
the global head of foreign exchange strategy at Deutsche Bank, published an
influential paper, Currencies:
Pensions’ Savior?, last August in which he
recommended a 20 to 30 percent allocation.
Since 1980, forex has exhibited long-term systematic returns that are comparable, if not better, than both bonds and equities. | Though Hafeez’s research was tailored to the needs of
institutional investors, his conclusions, if valid, would also apply to private
investors. "[Foreign exchange] has exhibited long-term systematic returns, or
beta, that are comparable, if not better, than both bonds and equities since
1980," he says. "It also has greater liquidity than both." Foreign exchange
returns also exhibit very low correlation with returns in the equity and bond
markets.
By long-term systematic returns—beta—Hafeez is referring to the
return attributed to market performance in general, rather than manager trading
or selection skill. Beta is often measured by a benchmark index for a given
market—the risk and return produced by the S&P 500, for example, reflects
the beta of large-cap U.S. equities. Investors who want to earn the
market’s beta return can therefore put their assets into products linked to the
S&P 500. However, it is far more complicated to capture the beta of
currencies. While it is easy to create a benchmark for broad-market returns in
equities—simply aggregate the performance of a given number of stocks, weighting
them by some criteria, typically market capitalization—creating a benchmark of
currency returns is another matter. Securities have an intrinsic value. But the
concept of value in foreign exchange is determined by how one currency moves in
relation to another. Investors can have long positions in equity or bond
markets, but in forex markets, to be long in one currency means, by definition,
to be short another. "If you have currency and you just sit on it, then you just
have risk in your portfolio," says Pierre Lequeux, head of currency management
at ABN Amro Asset Management in London. "You need to have an investment process
attached to currency in order to release the value of foreign exchange." There
is therefore no natural or easy way to create a beta benchmark for currency
returns. One work-around is to aggregate the performance of foreign
exchange-only fund managers. However, as with the hedge fund indices that adopt
a similar approach, this is fraught with a number of well-known problems, such
as the selection bias of managers, a lack of representation and survivorship
bias—the tendency for funds that have gone out of business to fall out of the
index, leaving only the strongest players and skewing the index unrealistically
upward. As an alternative, Hafeez suggests building an index using a
number of replicable rules. He would allow a currency strategy into this
benchmark if it meets several criteria. First, it must involve one of the
world’s top 10 currencies, in order to have a high degree of liquidity. It must
also fall into one of three of the most widely used forex investment styles: (1)
It must have a positive yield, also known as "carry"; (2) it must have positive
momentum; and (3) it must be undervalued. The specifics of how each individual
trade is put in place are complex, but these three strategies are the main ways
foreign exchange managers get value out of the market. When back-tested to 1980, these combined investment styles
deliver annualized returns of 12 percent, with a Sharpe ratio (a measure of
risk-adjusted return; the higher the ratio, the better the historical
risk-adjusted performance) of 0.92. The equivalent for equities is 11 percent
annual return and a Sharpe ratio of 0.27, and for bonds an average annual return
of 9 percent with a Sharpe ratio of 0.38. Meanwhile, equities and bonds have had
a 20 percent correlation in monthly returns since 1980, but the combined forex
strategies correlated just 2 percent with equities and -12 percent with bonds.
On paper, at least, this means foreign exchange is a good diversifier of
portfolio risk. The results led Hafeez to conclude that investors may be missing
opportunities to improve their portfolio performance by not having pure forex
exposure.
|