Professional Documents
Culture Documents
An Overview of Hedge Fund Strategies:, Including
An Overview of Hedge Fund Strategies:, Including
net/publication/251337307
Article
CITATION READS
1 8,296
2 authors, including:
Gregory Connor
National University of Ireland, Maynooth
60 PUBLICATIONS 2,281 CITATIONS
SEE PROFILE
All content following this page was uploaded by Gregory Connor on 13 October 2014.
This paper gives an overview of the most common hedge fund strategies, and their
implications for portfolio risk and return. The paper is intended as a review of the
existing literature rather than as an original contribution to the research frontier. The
paper should be accessible to masters-level students in finance, or investment
practitioners with good knowledge of modern finance theory. The first section looks
at some of the key issues in categorizing hedge fund investment strategies. Section
two discusses various long/short strategies; section three, event-driven strategies;
section four, tactical trading strategies; and section five, relative value strategies.
Section six discusses empirical methods used to identify a hedge fund’s strategy based
on its past returns, and section seven concludes. A bibliography and a glossary of
terms appear at the end of the paper.
In this section we discuss the key issues in defining hedge fund strategies and some of
broader strategy groupings that have been suggested. We will focus on specific, more
narrowly defined strategies in later sections.
Fung and Hsieh (1997) classify a hedge fund’s strategy according to both “style” and
“location”. Here, “style” refers to the type of positions the fund manager is taking,
such as going long and short, betting on a particular type of corporate event, or
maintaining market-neutrality. “Location” refers to the asset class that the hedge fund
invests in, such as fixed income, equity, or currencies.
∗
Connor is at London School of Economics, and Lasarte at Standard Life Investment Management.
We would like to thank Sheng Li and Abishek Kapur for helpful comments on an earlier draft.
1
See Connor and Woo (2003) for an introduction to hedge funds.
2
In fact, the cost of revealing likely trades is even larger than the lost profit opportunities. Competing
traders can “front-run” likely trades and thereby gain from the price impact associated with the other
manager’s trading behaviour. There is also the risk of a “liquidity squeeze” when the revealed trader
seeks to unwind his position.
1
Amenc, Martellini and Vaissié (2002) distinguish between “return enhancer”
strategies and “risk reducer” strategies. They cite distressed securities, event-driven
and macro funds as ‘return enhancers’, meaning that these funds seek very high
expected return but also increase overall portfolio volatility. On the other hand,
convertible arbitrage, fixed income arbitrage, long/short and short selling funds are
cited as ‘risk reducers,’ offering positive excess returns while also decreasing overall
portfolio volatility.3
Hedge funds are often segmented according to the asset class that they invest in.
There are hedge funds trading solely in equities; others specialize in fixed income or
currencies. Hedge funds can also be grouped according to the geographical location
of the assets they trade in (using the term “location” in a more traditional sense than
Fung and Hsieh above). For example, some funds may be focused on European assets
while others may be limited to emerging markets. A hedge fund that has a mandate to
invest in any country is labeled as a global fund.
Many funds are multi-strategy, where the hedge fund manager changes investment
strategies depending on market conditions or allocates capital across different
strategies simultaneously. A well-documented example is LTCM, which although
usually classified as a relative value fund in fact also took positions in merger and
takeover events.
One interesting division of hedge funds strategies is according to their historical roots.
Some hedge fund strategies trace their lineage back to the original hedge fund
industry, which was started by Alfred Winslow Jones in the late 1940’s. Others are
descendants of trading desk strategies originally practiced only within brokerage
houses and investment banks. Other strategies developed out of commodities and
futures trading (particularly but not exclusively currency trading). We will mention a
few of these historical roots when we discuss individual strategies later.
It should be noted that any attempt to establish a formal system of classification for
hedge fund strategies is limited by the fact that these strategies are continually
changing. As investment opportunities shift, hedge fund managers modify their
investment plans or design new ones to take advantage of profit opportunities. For
example, recently several hedge funds began taking direct positions in leveraged
buyouts, previously the domain of private equity firms.
A widespread method of hedge fund investment is the fund of hedge funds approach.
Funds of hedge funds allocate their capital by either investing in hedge funds with
different strategies, or investing in many hedge funds with the same strategy. We do
3
Here it is important to distinguish between the volatility of the hedge fund considered in isolation and
its effect on the total portfolio volatility of the final investor, who typically has only a small part of
total wealth invested in the fund.
2
not consider a fund-of-funds as a separate hedge fund “strategy” -- the strategy
classification is more appropriately applied at the individual fund level.
In the next four sections we will discuss some specific hedge fund strategies, grouping
them under four broad themes: long/short, event driven, tactical trading, and relative
value.
2. Long/Short Strategies
Long/short strategies exploit the ability of hedge fund managers to freely short
equities, an opportunity not available to most portfolio managers operating under
traditional investment mandates. Long/short strategies separate individual stock risk
from market risk.
Equity Market Neutral Strategy is the classic (and original) hedge fund strategy.
This was the strategy proposed by Alfred Winslow Jones, who started the first hedge
fund in 1949. A market neutral strategy uses the combination of buys and short-sales
(sometimes augmented by options and futures positions) to offset any correlation
between the portfolio return and the overall market return.
Alfred Winslow Jones’ motivation for using a market-neutral strategy was his
perceived information advantage: he considered himself a stock picker, rather than a
market timer. If a portfolio manager is not attempting to speculate on the overall
market move, then eliminating its ancillary effect on portfolio return serves to
highlight the contribution of the manager’s security selection ability to performance.
This also benefits the final investor in the hedge fund. If the final investor knows that
the hedge fund allocation will not affect his overall portfolio’s sensitivity to market
movements, then the contribution of the hedge fund to portfolio volatility is greatly
reduced.
Market neutral investing is related to the concept of portable alpha.4 Suppose that a
stock selector/fundamental analyst has superior information about the returns of a
collection of assets. In the context of a traditional active fund, this information can be
used to outperform the market benchmark by over-weighting those securities with
forecasts of positive returns and underweighting (or zero-weighting) those with
forecast negative returns. Alternatively, the stock selector/fundamental analyst can
exploit his superior information via a hedge fund. He takes long positions in the
securities he expects to have positive returns and short positions in securities where he
expects negative returns. Note that this hedge fund, combined with a benchmark
index tracking fund, is essentially identical to the same manager’s traditionally-
structured active fund. Due to his superior information, the stock
selector/fundamental analyst has “portable alpha” which can be implemented either
4
See Kung and Pohlman (2004) for a discussion of portable alpha.
3
via a hedge fund or via a traditional active fund. The market-neutral hedge fund
combined with a benchmark index tracking fund is equivalent to a traditional active
fund run by the same manager.5
By using a market-neutral hedge fund to exploit his superior information, the stock
selector/fundamental analyst allows final investors to make their own choices about
the market risk exposures of their aggregate portfolios. With the traditional active
fund, the choice of market risk exposure is bundled together with the stock selection
decisions. If the stock selector/fundamental analyst perceives no ability to add value
in terms of market timing, then it is better for him to leave this decision to final
investors and not dilute his overall performance with extraneous sources of return.
Some analysts have questioned the reliability of the “market neutrality” claim of
hedge funds so labelled. Patton (2004) differentiates between market neutrality
defined as zero correlation with the market return, and “complete neutrality” defined
as no dependence (including nonlinear dependencies) between the hedge fund return
and market index return. Complete neutrality accounts for options-like relationships
between the market return and hedge fund return. For example, the two returns might
be unrelated except when the market return is extremely large in magnitude. Patton
finds that empirically about a quarter of self-described market neutral hedge funds are
not in fact market neutral, using one or both of his two market-neutrality definitions.
Equity long/short is the same as equity market neutral except without any explicit
promise to maintain market neutrality. This increases the flexibility of the manager to
choose net-long or net-short (positive beta or negative beta) market exposure, while
still focussing primarily on stock-selection opportunities. It has the disadvantage
from the perspective of the final investor that it is no longer clear how the hedge fund
allocation affects overall portfolio risk.
Gatev, Goetzmann and Rouwenhorst (1999) analyse the increasing popularity of pairs
trading and attempt to simulate its risk-return pattern. Based on an algorithm
minimizing the sum of squared deviations between two prices, they identify pairs of
securities whose stocks have historically moved closely together. They then devise a
trading strategy that takes advantage of any deviation of the spread between the pairs.
Even accounting for transaction costs, the authors find that the strategy generates
significant excess returns.
5
See Connor and Woo (2003) for detailed analysis linking a hedge fund plus tracking index to a
traditional active fund.
4
Pairs trading is related to the more complex strategy called statistical arbitrage. A
pure arbitrage opportunity is a zero-cost portfolio with zero probability of a strictly
negative payoff and a positive probability of a strictly positive payoff. This is a “sure
thing - free lunch” and cannot exist in a well-functioning capital market.6 A statistical
arbitrage opportunity is a zero-cost portfolio where the probability of a negative
payoff is very small but not exactly zero. Ross (1976) was the original developer of
this concept and used it as the basis of his Arbitrage Pricing Theory. In practice, the
statistical arbitrage trader seeks to exploit a pricing discrepancy between related
securities while hedging against all pervasive sources of risk: market risk, sector risk
and factor-related risk (such as size, growth and value factors) and diversifying asset-
specific risk.
Dedicated Short Bias differs from the other long/short strategies in concentrating on
the short side and thereby sacrificing the market-neutrality feature. Not surprisingly,
this strategy tends to be most effective when markets are declining.
A motivation for dedicated short bias is the concept that, due to the restrictions on
short sales by most institutional investors, there are untapped profit opportunities
available to investors who can short-sell freely.7 Although explicitly not market-
neutral on its own, dedicated short bias can have the same diversifying potential as
traditional long/short, when it is part of an aggregate portfolio which includes long-
only investment mandates.
Usually, short sellers must focus on the shares of companies with large market
capitalisations, as they are more liquid and easier to borrow. Aggressive accounting
practices open up possibilities for short selling. The shares of Tyco, the US
conglomerate subject to an accounting scandal in 2002, were cited as an excellent
opportunity for short sellers who were able to read into the company’s
misrepresentations before the rest of the market caught on. In this way, short-sellers
can contribute to the informational efficiency and corporate monitoring role of capital
markets.
6
The non-existence of pure arbitrage opportunities is the most fundamental building block of modern
asset pricing theory. See Ross (1978) and Harrison and Kreps (1979) for some classic analysis.
7
See Diamond and Verechia (1987) for theoretical analysis of short-sale constraints and their effects
on market informational efficiency.
5
counterparts. Hedge funds following a distressed security investment strategy are not
limited to a particular asset type. They are active in bonds, stocks, bank debt, trade
claims, private placements, warrants, etc. Some hedge funds focusing on distressed
securities are active in the entire market; others limit themselves to more specific
sectors. For example, some hedge funds may concentrate solely on the
telecommunications sector.
This technique attempts to profit from both relative and absolute pricing inefficiencies.
The market for distressed securities is often inefficient and illiquid because of factors
such as investor irrationality, risk aversion, legal restrictions on holding sub par
securities, low coverage by analysts, and lack of research. An active form of this
strategy is to buy a substantial proportion of the outstanding security of the distressed
company and then attempt to influence the restructuring process. The distressed
security strategy generally involves taking long positions, although a hedge fund
manager could take short positions in distressed securities he thinks will worsen.
The market for distressed security grew in the 1980s in the US as the number of
leveraged buyouts and hostile takeovers increased. In Europe, the market is smaller.
One of the reasons is that bankruptcy laws in European countries are less transparent
than in the US. Japan is a growing market for distressed debt.
Merger arbitrage carries a variety of risks. When there are antitrust issues concerning
the merger or takeover, there is the possibility that regulators will block the deal.
There is also a financing risk, in that the acquirer could lose the financial backing to
carry out the purchase. Some hedge funds invest in a large number of deals
simultaneously to hedge their risks. Others take a more concentrated approach.
6
The volume of mergers and takeovers taking place tends to be highly cyclical, which
influences the opportunities for merger arbitrage. During “slow” periods the available
profit opportunities are correspondingly small.
Tactical trading strategies attempt to profit by forecasting the overall direction of the
market or a market component. The payoffs of hedge funds specializing in this area
depend on how well a hedge fund manager can forecast price movements as well as
predict the exact timing of these movements.
Macro strategies make large bets based on forecasts of major macroeconomic events
such as changes in interest rates, currency movements and stock market performance.
In searching for investment opportunities, strategists will take into consideration a
diverse set of factors such as geopolitical issues, economic indicators, market trends
and liquidity flows. Macro funds are active in a wide variety of markets including
foreign exchange, futures, fixed income, equity, as well as both exchange-traded and
over-the-counter securities.
Macro funds often take on leverage and actively use derivatives. Managers have
substantial flexibility and invest in any country or asset class where they see an
opportunity. The strategy relies on the ability to make superior forecasts compared to
other market participants and then act swiftly and decisively.
The return profile of macro funds is much more volatile than that of other hedge funds.
Part of the reason is that macro funds often trade in instruments that are relatively
illiquid. Also, correlation among assets in emerging markets, where these funds often
invest, is much higher than in developed markets. There are a unique set of risks in
investing in emerging markets such as political hazards and the lack of an adequate
legal system.
An example of the type of investment followed by macro funds is the yen carry trade.
At its peak in mid-1998, de Brouwer estimates that there was $200-$300 billion
invested in this strategy, half of it coming from hedge funds. The trade consisted of
borrowing Japanese yen, taking advantage of extremely low interest rates in Japan,
and investing the proceeds in dollar-denominated assets such as Treasury bills, which
7
had higher yields. As long as the yen did not appreciate, the strategy provided
substantial payoffs.
As with the long/short investing approach, this class of strategies involves buying and
selling two or more related securities. The risk of these strategies depends upon how
closely related are the securities bought and sold short. Generally this strategy is
characterized by low volatility and little correlation with the market.
It is common for hedge funds active in relative value strategies to hedge their
exposure to the price movements of the underlying securities, interest rates, and broad
market movements. Since relative value price discrepancies tend to be small, hedge
funds operating in this area tend to be among the most leveraged, in order to magnify
potential gains.
8
Liang (2004) however argues that CTAs follow distinctly different trading strategies from hedge
funds.
8
senior to equity in the case of bankruptcy, and so its value depends on interest rates.
On the other hand, it can be analysed as equity in that the holder has the right to
convert into a fixed number of common shares. And finally, the security carries an
implicit option to exchange the bond for shares. There are often call provisions,
where the issuer has the right to buy back the issues, or put provisions, in that holder
can redeem the issue at a fixed price. If the share price rises, a convertible bond price
will tend to behave similarly to a share price. If the share price falls, convertible
prices are more closely related to those of regular bonds.
Convertible arbitrage strategies typically take a long position in the convertible bond
and short the company’s equity. In doing so, the investor takes advantage of the
undervaluation of the convertible bond while reducing the exposure to the underlying
stock price movement. An important parameter in this trade is delta, the proportional
change in the value of the convertible bond relative to the change in the underlying
stock price. If the stock price increases substantially, convertible bond prices
generally increase even more rapidly. The interest on the proceeds from selling short
the underlying stock is used to finance the position.
Convertible arbitrage strategies are subject to equity risk, interest rate risk and credit
risk (since the bond component of convertibles tends to be low grade). Agarwal,
Fung, Loon and Naik (2004) find, somewhat surprisingly, that convertible arbitrage
funds active in the Japanese market only show significant exposures to equity market
risk, not to interest rate or credit risk factors. They conjecture that their findings may
not generalize to other convertible bond markets.
Hedge funds are now one of the main sources of liquidity in the convertible bond
market. Lhabitat (2002) notes that hedge funds are estimated to own 30% of all
convertible bonds outstanding in the US.
Amazon, the online retailer, offers a useful example of how convertible arbitrage
strategies can provide payoffs independently of the general market trend. In 1999, the
company had a large tranch of outstanding convertible bonds due to mature in 2009.
At the time, the company’s shares were priced under the assumption of
extraordinarily high future growth. However, Amazon’s convertible bonds were
greatly undervalued. Arbitrageurs bought these securities and shorted Amazon’s
9
stock, while hedging the credit risk. In the following year, even as the price of
internet stocks fell, the convertible arbitrage trade yielded high returns as the spread
between the valuation of Amazon shares and those of its convertible bonds fell
substantially.
Equity and debt markets often react to new information differently. For example, a
poor earnings announcement might affect a company’s share price immediately but
not the price of its debt. Unlike other relative value strategies, capital arbitrage
involves taking positions in two different asset classes. In devising a strategy, a hedge
fund manager needs to study the correlation between a company’s debt and equity. A
more complex example of capital arbitrage involves taking positions in derivatives
dependent on one of the assets, such as high yield debt and call options on the stock.
An example of this type of trade would be the case of British Telecom (BT). The
company’s debt was downgraded several times during 2001. BT’s bond prices
plummeted; however, its share price was somewhat slower to react. Arbitrageurs took
advantage of the time lag by taking long positions on BT’s debt and shorting the
equity.
Fixed income arbitrage strategies attempt to exploit mispricing among fixed income
securities. This strategy relies heavily on mathematical models of the term structure
of interest rates to identify mispricing and manage positions. There are a number of
different techniques that can be identified within the fixed income arbitrage strategy,
and some of these are discussed below. Most hedge fund managers engaged in fixed
income arbitrage will use several of the techniques, depending upon perceived
opportunities. Fixed income arbitrage has its historical roots in the fixed income
trading desks of brokerage houses and investment banks.
Yield curve arbitrage involves taking offsetting positions at various points of the
yield curve, typically those of a Treasury bill, in response to perceived disequilibrium
price relationships. This strategy also can involve looking for pricing imperfections
between the futures and cash treasury markets. Hedge funds active in this area often
take positions in the futures and swaps markets as an alternative to directly buying or
selling bonds.
Consider the following simple scenario. Suppose that the 5-year Treasury bill has a
current yield which is substantially above the yield of the 4-year and 6-year bills.
Suppose that there is no obvious tax or call provision in the bond which explains this
price differential. A yield curve arbitrage strategy would be to buy the relatively
10
cheap 5-year Treasury bill while shorting the 4-year and 6-year bills. Whatever
temporary liquidity pressure has caused the 5-year to be underpriced relative to the 4
and 6-year bills is likely to disappear over time, allowing the manager to unwind the
position at a profit. The manager is hedged against general interest rate moves since
the 5-year interest rate sensitivity is offset by the short position in the 4 and 6-year
bills.
An actual example of a recent fixed income arbitrage trade took place in the first half
of 2000. At the time, there were forecasts of a diminished supply of long-term
Treasury bonds because of US budget surpluses. This created an excess demand for
longer-term issues. As a result, the yield on the 30 year issue was lower than the yield
on the 10 year bond, a rare event given that the yield curve is usually upward sloping.
Hedge funds took short positions in the shorter-dated bonds, and long positions in the
longer-dated securities. Managers were able to unwind at a profit when the normal
shape of the term structure was re-established.
Corporate spread arbitrage involves making a bet on the evolution of the interest
rate spread between the prices of corporate bonds and Treasury bills. Depending on
whether the arbitrageur believes the spread will increase or tighten, the strategy
involves taking long and short positions in the two securities. An assessment of the
repayment capacity of the corporation issuing the bond can give some indication of
where the spread will move. Several authors claim there is some forecasting potential
in corporate spreads. A very similar strategy involves taking positions in emerging
market debt instead of corporate bonds.
A closely related trade was conducted by LTCM in 1994 using Italian securities. The
anticipation of the European monetary union and a lack of trust in the Italian
government’s fiscal policy resulted in a yield on domestic-currency denominated
Italian government bonds higher than the Italian lira swap spread in the London
interbank market (an agreement to exchange a fixed interest rate for the floating rate).
LTCM took advantage of this differential by borrowing at a floating interest rate to
buy Italian government bonds. LTCM then repaid the floating rate loan by entering
into an interest rate swap, exchanging the fixed payment from the government bond
for the floating rate. Given that the fixed amount necessary to carry out this trade was
lower than the coupons provided by the government bond (because of a higher yield
than the swap spread), LTCM pocketed the difference.
11
Mortgage-backed securities are very difficult to price given that homeowners have the
option to refinance and repay their loans early. A common trade is to take a long
position on a mortgage-backed security and to short a government bond of similar
maturity. That way, the investor receives a payoff from the yield differential while
hedging the risk from interest rate surprises.
In the early 1990s, mortgage arbitrage was one of the most profitable trades for
LTCM. LTCM’s traders took long positions in these securities, and short-sold
Treasury bonds of similar maturity to hedge interest rate risk. In addition, LTCM
used interest rate swaps to hedge for the risk of the yield curve shifting, and bought
Treasury bond options to replicate the options implicit in the mortgage-backed
securities.
Stock index futures arbitrage takes advantage of the spread between index futures
and price of underlying securities. If the futures price is lower, the manager goes long
the future contract while shorting the individual component stocks (or enough of them
to create a replicating basket) and waits for the price differential to disappear, and
then unwinds the positions at a profit. The opposite arbitrage strategy applies if the
futures price is higher than the index-implied price.
ADR arbitrage exploits the difference between the trade prices of American
depositary receipts (ADRs) and underlying shares. ADRs are securities issued and
traded in a US market such as the New York Stock Exchange and priced in dollars,
but backed by foreign-traded shares of a foreign-domiciled company. Sometimes the
currency-translated price of the foreign security traded in its home market differs from
the ADR price traded on the US market in dollars. Speculators able to act quickly can
attempt to exploit these arbitrage price differentials.
Most hedge fund databases classify hedge funds into strategy groups based solely on
self-identification. Time-series analysis is helpful in identifying the risk and return
profile of individual hedge funds and hedge fund indices from actual return histories
rather than self-reported strategies.
Schneeweis and Spurgin (1996) attempt to explain hedge fund returns based on the
type of assets traded and the trading style of the hedge fund managers. They use a
broad set of financial and economic indicators, including equity index returns, interest
rates, exchange rates and commodity prices, as well as the volatility of these assets
and a moving average of commodity and financial futures contract returns. They find
that hedge funds are driven by a different set of factors than traditional bond and
equity funds.
12
Martin (2000) also uses this method but first conducts cluster analysis, to group hedge
funds according to their particular strategy. After classifying each hedge fund, Martin
uses regression analysis to establish a link between the performance of each strategy
and a set of economic factors: interest rate swaps, aggregate bond prices, emerging
market bond prices, the return on the S&P 500 and the implied volatility on S&P 100
options. Martin finds there is a significant correlation when each strategy is grouped
as an index. When the analysis is applied to individual funds the coefficients vary
across funds and are not always significant.
Several problems are encountered in applying style analysis to hedge fund returns.
First, with hedge funds it is not always clear what are the appropriate market indices
to include as explanatory variables. Second, hedge funds often use dynamic trading
strategies that switch asset exposures rapidly, and this invalidates the standard
regression model. For traditional funds, exposure to an asset class is relatively static
and so time-series regression is less problematic. Third, Sharpe’s coefficient
restrictions (all coefficients nonnegative and summing to one) are not appropriate for
most hedge funds. For example, a long-short equity fund could have a positive,
negative, or zero exposure to the equity market index, depending upon whether it has
under-hedged, over-hedged, or exactly hedged its underlying market positions. No a
priori sign restriction on the coefficient is appropriate. Similarly, leverage and the use
of derivatives means that the sum of a hedge fund’s exposures can easily exceed one.
Fung and Hsieh (1997) modify and extend Sharpe’s style analysis to improve its
applicability to hedge funds. They attempt to capture the non-linearity displayed in
hedge fund returns by including non-linear regressors with options-like payouts.
They allocate the hedge fund’s returns to three components: the returns of the asset
class traded by the fund (“location”), the type of trading strategies followed by the
fund, and the amount of leverage used. Fung and Hsieh use a multistep approach.
First, they identify five mutually orthogonal components in hedge fund return data
using principal component analysis. Second, they construct portfolios of hedge funds
correlated with each principal component. They find five different strategies explain
much of the data: technically driven strategies which follow trends, opportunistic
strategies which place bets on market movements, macro funds, buyers of
undervalued securities, and distressed security funds. Third, they then perform a style
13
analysis for each of the five hedge fund strategy portfolios, regressing returns to
traditional asset benchmarks. Fourth, to identify dynamic trading strategies, Fung and
Hsieh introduce a technique in which the returns of a hedge fund strategy are analysed
in different market environments. The technique involves listing quintiles of returns
for one of the different asset classes (equities, government bonds, etc). Hedge fund
strategy returns are analysed in each quintile. If a strategy uses a buy-and-hold
approach, the returns in the five states of the world should be closely aligned to the
asset class. On the other hand, returns that are non-linear and especially large in
market extremes are taken as evidence that a hedge fund strategy is following a
dynamic trading strategy.9 These returns can only be replicated through securities
offering non-linear payoffs such as derivatives.
In their empirical analysis, the Fung and Hsieh find that trend-following hedge funds
have a return profile similar to a straddle position in US equities. During their sample
period the payoffs of opportunistic funds are similar to a call option on gold. Global
macro funds had a similar payoff to holding a straddle on the dollar. They construct a
model consisting of nine long positions on the traditional assets plus three dynamic
trading strategies which can characterize a “substantial part” of the trading strategies
of hedge funds.
Agarwal and Naik (2000b) also use modified style analysis to analyse hedge fund
strategies. As explanatory variables they construct portfolio returns that include
options, to better track the nonlinearities in hedge fund returns. Like Fung and Hsieh
they recognize three sources of hedge fund performance: “location”, trading strategy
and leverage. They utilize equity, commodity and currency indices to reflect the
location element of an investment strategy and as well a collection of zero-net-
investment strategies (the difference between small firm and big firms returns, interest
rate spreads, etc). To capture the trading strategy element, they use a basket of in-the-
money European options (present value of exercise price equals current index value),
out-of-the-money options (exercise price is half a standard deviation away from the
index value), and a deep out-of-the-money options (exercise price is one standard
deviation away from the index value) on several indices. They find that “location”
and strategy have useful content, but the difference in excess returns between
leveraged and non-leveraged funds is statistically insignificant.
Mitchell and Pulvino (2000) adopt a similar approach. Looking solely at those hedge
funds specializing in merger arbitrage, the authors run a regression of hedge fund
returns on an equity market index with dummy variables indicating whether returns
were above or below a negative 4% threshold level. In flat or rising markets, the
average market exposure (beta) of the funds is essentially zero. However, when the
market experiences a monthly decrease of over 4%, merger arbitrage funds are
positively correlated with the market return. They draw a parallel between
investments in the merger arbitrage strategy and holding a long position in a risk-free
bond and a short position in a stock index put option.
9
Those assets are MSCI US equities, MSCI non-US equities, IFC emerging market equities, JP
Morgan US government bonds and JP Morgan non-US government bonds, the one-month eurodollar
deposit rate, a trade-weighted dollar exchange rate and the price of gold and a high-yield bond index.
14
Instead of using option-type returns to capture the non-linearity inherent in hedge
fund returns, Lhabitant (2003) uses hedge fund indices. The author uses the nine
CSFB/Tremont hedge fund indices in his study. Lhabitant also proposes a rolling
window regression to capture changes in hedge fund strategies.
7. Conclusion
Hedge fund strategies are extremely varied. Given this wide diversity, it is
appropriate to ask whether there is any common feature to the class of strategies
practiced by hedge funds. The surprising answer is yes: there is a common theme
linking the strategies. Hedge funds focus on strategies that traditional funds cannot
undertake, or cannot do well. The three defining characteristics of hedge funds are
very strong incentive structures for managers, light regulation, and investment opacity.
Hedge funds undertake investment strategies which play to one or more of these
strengths. Long-short strategies rely on hedge funds’ freedom to short equities and
take derivatives positions, both of which are often forbidden for traditional funds, or
restricted by funds’ investment mandates. Hedge fund strategies requiring razor-
sharp decisions in risky environments, such as event-driven strategies and tactical
trading strategies, are helped by the very strong management incentives used by
hedge funds. The investment opacity of hedge funds is crucial in many relative-value
strategies since it prevents other investors from uncovering the fund’s positions and
squeezing its transactions in illiquid markets.
15
Table 1: Hedge Fund Strategy Classifications
16
Glossary
Absolute Return—Portfolio return without subtracting any benchmark return.
Active Risk—Standard deviation of active return. The term is also sometimes used to
refer to the difference between the risk exposures of the portfolio and the benchmark.
ADR—Securities issued and traded in a US market and priced in dollars, but backed
by foreign-traded shares of a foreign-domiciled company,
Arbitrage Pricing Theory—An asset pricing model that specifies returns as a linear
function of multiple factors in an arbitrage free economy.
17
Complete market neutrality—A fund with no dependence (including non-linear
dependencies) between its return and the market index return. This is stronger than
no correlation which covers only linear dependence.
Credit spreads—Additional yield offered by a risky bond over ‘riskless’ bond yield to
cover for risk of default by issuer.
Delta—The change in the value of a derivative security relative to the change in the
underlying asset price.
18
Equity long/short—Same as equity market neutral except without any explicit
promise to maintain market neutrality (see Market Neutral).
Global Macro—Hedge fund strategy where large directional bets are made, often on
the direction of currency exchange rates or interest rates.
Hedge Fund Research (HFR)—Hedge fund index provider (for details on strategies
covered see Table 1 in the text).
19
Mortgage-Backed Securities (MBS) —Pools of mortgage-backed debt developed by
financial institutions.
Pairs trading—Strategy consisting of the combined purchase and sale of two similar
securities to exploit relative mispricing.
Passive Returns—Returns from holding a benchmark, such as the S&P 500 or MSCI
EAFE.
Portable alpha—The concept that superior stock selection ability can be expressed via
a variety of different investment vehicles: such as long-only versus long-short funds.
(See Alpha).
20
Style analysis—A particular application of regression analysis of fund returns in
which the explanatory variables are returns on a set of market indices, the regression
coefficients must be non-negative, and the sum of the coefficients must add up to one.
TASS Research—Hedge fund index provider (for details on strategies covered see
Table 1 in the text).
VaR (Value at Risk)—The maximum loss to a portfolio over a given time period with
a given level of confidence. For example, if a 10 day VaR at 99% confidence level is
$100,000, then we conclude that 99% of the time the portfolio will not decline more
than $100,000 in value within 10 days.
21
Bibliography
Amenc, Noel, Lionel Martellini and Mathieu Vaissi (2002), “Benefits and Risks of
Alternative Investment Strategies,” working paper, Edhec Business School, Lille.
Agarwal, Vikas and Narayan Y. Naik (2004), “Risk and Portfolio Decisions Involving
Hedge Funds,” Review of Financial Studies 17, 63 – 98.
Agarwal, Vikas, Narayan Y. Naik, William Fung and Yee Cheng Loon (2004), “Risks
in Hedge Fund Strategies: The Case of Convertible Arbitrage,” working paper,
Georgia State University.
Connor, Gregory and Mason Woo (2003), “An Introduction to Hedge Funds,”
working paper, Financial Markets Group, London School of Economics.
Currie, Anthony and Jennifer Morris (2000), “And Now for Capital Structure
Arbitrage,” Euromoney Magazine, December.
Fung, W., and D.A. Hsieh (1997) "Empirical Characteristics of Dynamic Trading
Strategies: The Case of Hedge Funds," Review of Financial Studies, 10, 275-302.
Fung, W., and D.A. Hsieh (2001) “The Risk of Hedge Fund Strategies: Theory and
Evidence from Trend Followers,” Review of Financial Studies, 41 313-341.
Gatev, Evan G., William N. Goetzmann and K. Geert Rouwenhorst (1999), “Pairs
Trading: Performance of a Relative Value Arbitrage Rule,” working paper, School of
Management, Yale University.
Harrison, J.M., and D.M. Kreps (1979), “Martingales and Arbitrage in Multiperiod
Securities Markets,” Journal of Economic Theory 20, 381-408.
22
Kung, Edward and Larry Pohlman (2004), “Portable Alpha,” Journal of Portfolio
Management, 30:3.
Lhabitant, François-Serge (2002), Hedge Funds: Myths and Limits, John Wiley &
Sons, New York, USA.
Liang, Bing (2004), “On the Performance of Alternative Investments: CTAs, Hedge
Funds and Funds-of-Funds,” forthcoming in Journal of Investment Management.
Martin, George (2000), “Making Sense of Hedge Fund Returns: What Matters and
What Doesn’t,” Derivatives Strategy.
Mitchell, Mark, and Todd Pulvino (2001), “Characteristics of Risk in Risk Arbitrage,”
Journal of Finance 56, 2067-2109.
MSCI (2003), Investable Hedge Fund Indices Methodology, Morgan Stanley Capital
International Inc.
Okunev, John and Derek White (2003), “Hedge Fund Risk Factors and Value at Risk
of Credit Trading Strategies,” working paper, School of Banking and Finance,
University of New South Wales.
Patton, Andrew (2004), “Are "Market Neutral" Hedge Funds Really Market Neutral?”
working paper, Financial Markets Group, London School of Economics.
Ross, S.A. (1976), “The Arbitrage Theory of Capital Asset Pricing,” Journal of
Economic Theory 13, 341-360.
Ross, S.A. (1978), “A Simple Approach to the Valuation of Risky Streams,” Journal
of Business 51, 453-475.
Standard & Poor’s (2004), Standard & Poor’s Hedge Fund Index: Structure,
Methodology, Definitions, and Practices, Standard & Poor’s Inc.
23