Full text of Economic Review (Federal Reserve Bank of Atlanta) : Fourth Quarter 2006, Volume 91, Number 4

View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Preface—Hedge Funds:
Creators of Risk?
GERALD P. DWYER JR.
The author is a vice president and head of the financial team in the Atlanta Fed’s research
department. This preface provides an overview of the Atlanta Fed’s 2006 Financial Markets
Conference, “Hedge Funds: Creators of Risk?” held May 15–18.

s luck would have it, the Federal Reserve Bank of Atlanta’s 2006 annual financial
markets conference focused on hedge funds just as such funds became the subject
of numerous news articles and discussions at regulatory agencies. The conference
was held in May, when registration of hedge funds recently had become required, the
amounts flowing into hedge funds was mushrooming, and many people were wondering when new regulations were to follow.
It may seem like a question with an obvious answer, but what is a hedge fund anyway? Many, including me before the conference, would answer that a hedge fund is
similar to a mutual fund except that it may accept investments only by relatively well
off people. More precisely, only investors who have more than a million dollars in
assets or earn more than $200,000 per year may invest in hedge funds. In fact, thinking of hedge funds as being particularly similar to mutual funds is not a useful way to
think about hedge funds.
As William Fung and David Hsieh show in their paper, hedge funds are far more
than mutual funds with some complex financial strategies. Hedge funds specialize in
buying and selling numerous kinds of risks, a fact that is well known in the industry
but is not widely known outside it. In their very interesting paper in this issue of the
Economic Review, Fung and Hsieh go far beyond dispelling this common misperception. They discuss a substantial amount of data on the hedge fund industry and
provide an informative discussion of the economics of the industry.
Arguably, one reason that hedge funds have received so much attention is participation by more and more people. One reason more people are using them is
because, when the definition of accredited investors as those with substantial wealth
or high incomes was written, a million dollars was a substantial amount. But a million
dollars today is not what it used to be. In the 1950s, on the television drama “The
Millionaire,” everyday people received a million dollars tax free from an anonymous
donor. The show’s premise was that the recipient was very rich after this gift.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Watching lotteries today, it clearly takes quite a few millions before someone can regard
herself as having become rich overnight.
Franklin Edwards discusses the rationales for allowing only relatively wealthy
people to participate in hedge funds. He suggests that, rather than limiting participation in hedge funds, the public would be better served by removing some of the
regulatory restrictions on the financial strategies available to mutual funds.
In their paper, Nicholas Chan, Mila Getmansky, Shane M. Haas, and Andrew Lo
note that many banks now operate proprietary trading units that are organized much
like hedge funds. As a result, the hedge
fund industry’s risk exposures may signifiHedge funds specialize in buying and sellcantly affect the banking sector, creating
ing numerous kinds of risks, a fact that
new sources of systemic risk. To quantify
is well known in the industry but is not
this potential impact on systemic risk, the
researchers develop several new risk meawidely known outside it.
sures for hedge funds and apply them to
individual and index data. They interpret their provocative findings as suggesting
that hedge funds may be heading into a period of lower expected returns with systemic risk on the rise.
Hedge funds have been criticized for the compensation received by the funds’
managers, which seems quite high compared to that for mutual fund managers. Hedge
funds’ role in the market for corporate control has also been criticized, with a common interpretation effectively likening hedge funds to bandits who come in, lay off
employees, and sell the corporation’s assets, all in the name of making a quick buck.
Bruce Lehmann argues that little will be learned about hedge funds’ governance
or their role in the market for corporate control by looking at mutual funds. Instead,
he suggests that more can be learned by comparing hedge funds to firms with similar
assets and liabilities. Starting from this premise, he proposes that many hedge funds
can usefully be compared to proprietary trading desks at investment banks. In that
regard, he notes that the much-criticized compensation schemes at hedge funds
compare reasonably well with the schemes used in that environment. He also argues
that hedge funds’ efforts to improve corporate government benefit all shareholders
unless the target firm pays the hedge fund an outsized payment—greenmail—at the
expense of other shareholders.
All in all, the conference—as the papers in this issue of the Economic Review
demonstrate—provided a substantial amount of information and thoughtful analysis
on a little understood industry.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Hedge Funds: An Industry in
Its Adolescence
WILLIAM K.H. FUNG AND DAVID A. HSIEH
Fung is a visiting professor of finance at the BNP Paribas Hedge Fund Centre at the London
Business School. Hsieh is a professor of finance at Duke University’s Fuqua School of
Business. This paper was presented at the Atlanta Fed’s 2006 Financial Markets
Conference, “Hedge Funds: Creators of Risk?” held May 15–18.

t could be said that the hedge fund industry, compared to its brethren in the
asset management arena, was in its infancy up until a decade ago. Information
about these funds, both qualitative and quantitative, was not freely available to the
general investment public until academic research on hedge funds started in the
1990s, with Fung and Hsieh (1997), Eichengreen et al. (1998), Schneeweis and Spurgin
(1998), Ackermann, McEnally, and Ravenscraft (1999), and Brown, Goetzmann, and
Ibbotson (1999).
At the turn of the century, coinciding with the bursting of the Internet bubble,
institutional investors began increasing their allocation to hedge funds, responding in
part to the lackluster performance of global equity markets. As a result, assets under
management (AUM) by the hedge fund industry grew exponentially, and the number
of hedge funds doubled over the past five years, by some estimates.1 Consequently,
institutional investors figure more prominently in the industry’s clientele. This clientele shift in turn precipitated profound changes in the way hedge funds operate—
such as increased transparency, better compliance, and higher operational standard,
to name just a few. Some have referred to this change as the “institutionalization” of
the hedge fund industry.
Accompanying these changes came the rising demand for rigorous research into
hedge fund performance. At the same time publicly available hedge fund databases
became ubiquitous. Together they have spawned a fast-growing body of published
studies on hedge funds—professional as well as academic. Although there is no official
count of academic papers on hedge funds, a reasonable conjecture is that their numbers have grown at an even faster pace than the hedge fund industry.
This paper provides an overview, albeit somewhat biased, on a particular school
of thought in this growing body of hedge fund research. The school of thought to
which we refer is the thesis put forward in Fung and Hsieh (1999) on the economic
rationale of the hedge fund business model:

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Consider the problem confronting a money manager who believes that he has a
set of skills that could earn above average risk adjusted returns. We are not
advocating the existence of such strategies, but merely the hypothesis that the
manager believes this to be so. Let us assume that the manager has a limited
amount of personal wealth. In order to meet the fixed costs of a trading operation, the manager must leverage his skills and beliefs by attracting external
capital. Basically, he is financing a new venture. The choice is either equity
financing, in the form of a fund, or debt financing, in the form of putting up personal assets as collateral against borrowed capital. In most cases, the manager’s
personal wealth is insufficient to secure sizeable debt financing. That leaves the
formation of a fund as the only practical financing option. (1999, 317; italics
added for emphasis)

This simple characterization of the hedge fund business model points to a fundamental question frequently found in hedge fund research over the past few years. In
order for an opaque “new venture” to prevail in charging investors hefty incentive
fees, it must offer returns that are not easily available from more conventional, lowercost alternatives such as mutual funds. Putting aside the implausible hypothesis that
superior hedge fund returns reflect the “free lunch” that has escaped other investment
professionals, we face some basic questions: How is superior performance generated?
What are the risks? Can the superior performance last? Satisfactory answers to these
questions must begin with a clear understanding of the systematic risk factors inherent in hedge funds strategies.
These, in our view, are the foundational questions to be addressed in hedge fund
research. From an investor’s perspective, the answers are the key determinant of
whether hedge funds provide diversification to a portfolio of conventional assets
and—more important—whether hedge funds offer returns commensurate with the
fees they charge on a risk-adjusted basis. From the perspective of a counterparty to
hedge funds—such as prime brokers, commercial banks, and investment banks—
these answers are the key input for assessing the capital at risk for engaging in servicing and financing hedge fund businesses. From the regulator’s perspective, these
questions are key in the monitoring of the convergence risk of highly leveraged opinions that can destabilize markets, creating systemic risk.
Ultimately, hedge fund managers are guided by the desire to maximize the enterprise value of their firms. Like most other investment opportunities, different hedge
fund strategies must yield to constraints such as diminishing return to scale (capacity
issues) as well as other unrelenting forces of economic cycles (such as strategies
falling in and out of favor, often at the mercy of market forces). Rational choices within
the hedge fund business model—such as the degree of leverage, the allocation of risk
capital to factor-related bets versus delivering alpha (excess return above exposure to
risk factors) to investors—must in turn depend on the compensation contract between
the hedge fund manager and investors (“the fee structure”). Logically, optimal contracting between investors and hedge fund managers (and, for that matter, between
prime brokers and hedge fund managers) must take into account the presence of systematic risk factors inherent in hedge fund strategies.
In times of market stress, investors tend to take flight to liquidity. The transmission mechanism leading to a systemic withdrawal of risk capital from markets is not
necessarily driven by market prices. An equally plausible proposition is that when
faced with inadequate transparency, investors are innately unwilling to absorb performance shocks. Put differently, when opaque investment vehicles perform poorly,

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

it is hard for investors to differentiate between random shocks and systemic adverse
causes. Therefore, it is in everyone’s interest—hedge fund investors, financial intermediaries who provide leverage to hedge funds, and the hedge fund managers themselves—to identify the appropriate level of disclosure so as to avoid the risk of a
boom-bust capital formation process in the hedge fund industry. It may be impractical for hedge fund managers to publicize details of their trading positions. However,
it is important to identify the risk factors that underlie different hedge fund strategies in a way that helps investors assess
Are hedge fund managers evil geniuses who
the impact of changing market conditions
on hedge fund styles.
profit from wreaking havoc in capital marFinally, better compensation contract
kets, or do they represent “smart money,”
design reduces the risk of performance
providing risk capital to capital markets
shock, helps smooth the capital formation
process of the hedge fund industry, and
unencumbered by securities regulations?
ultimately enhances the enterprise value
of the hedge fund firm. Properly structured, compensation contract design can provide another important deterrent to hedge fund managers engaging in excessive
leverage that may otherwise be encouraged by a loosely specified incentive contract.
Together with the institutionalization of the hedge fund industry, the desire to
enhance the enterprise value of the hedge fund firm may dissuade—one can hope—
hedge fund managers from applying excessive leverage. This development in turn
will better align the interests of hedge fund managers, investors, prime brokers, and
regulators. We postulate that a necessary condition for better contract design is the
recognition and proper measurement of systematic risk factors inherent in hedge
fund strategies. It stands to reason that investors seeking alpha are likely to price the
services of hedge fund managers differently from those seeking leveraged factor bets.
To begin our discussion, we provide some recent statistics on the hedge fund
industry, particularly with respect to the growth in the size of the industry and the
emergence of a dominant institutional clientele among hedge fund investors. Here we
report empirical evidence showing that investors seeking alphas allocate capital differently from investors seeking factor bets.
An important prerequisite to estimating systematic risk factors in hedge fund
strategies is to have an accurate performance history of the hedge fund industry. We
discuss the problems with biases in hedge fund databases that must be recognized to
obtain accurate measures of returns.
Our paper then summarizes the research by addressing the fundamental question
of systematic risk exposure. We follow the framework in Fung and Hsieh (1997),
modeling hedge fund returns as a function of three key elements—how they trade,
where they trade, and how the positions are financed. The answers to the questions
of how and where hedge funds do their business are based on extensions of the
approach used to model conventional investment vehicles such as mutual funds (see,
for example, Sharpe 1992). Here, we examine the systematic risk factors in a number
of different hedge fund strategies.
The question of how hedge fund positions are financed brings up several unconventional and important issues peculiar to the hedge fund business model. First and
foremost, the ability to leverage, coupled with the existence of common risk factors
among different hedge fund strategies, raises the question of market impact. Put
1. Hedge Fund Research (HFR) estimated the number of hedge funds to be 3,617 in 1999 versus the
latest estimate of 8,219 as of the end of June 2005.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

differently, what if most hedge fund managers agree, albeit for different reasons, on
the “best trade”? Given that hedge fund bets are generally leveraged, what are the
risks of another Long-Term Capital Management (LTCM) incident?
What does academic research have to say on the following question: Are hedge
fund managers evil geniuses who profit from wreaking havoc in capital markets, or do
they represent the quintessence of “smart money,” providing risk capital to capital
markets unencumbered by securities regulations? The paper provides a brief summary of the empirical findings on this question.
In light of the clientele shift among investors in the hedge fund industry, we report
recent findings on the return experience of alpha-oriented investors. Here we refer to
the recent paper by Fung et al. (2006). The empirical evidence points to a declining
trend of alpha to investors. This trend is consistent with the implication of the Berk
and Green (2004) model. Demand growth for alpha, coupled with the layers of fees
charged by hedge fund managers and funds-of-hedge-fund managers, has led to a disproportionate share of returns in favor of product providers at the cost of investors.
New research on synthetic hedge funds (replication of hedge-fund-like returns via
mathematical models) at a lower cost to investors has been put forward. However, we
are not persuaded that the creation of synthetic hedge funds is a realistic solution to
the supply-demand imbalance between alpha producers and alpha buyers. The price
for alpha (implicit in the fee structure) has to be set so that it will encourage an
increase in alpha production. Ultimately, better alignment of interest between hedge
fund managers and investors through more appropriate compensation contracts must
be addressed. A necessary condition for better contract design is the identification of
non-alpha-related factor bets inherent in hedge fund returns. In the paper’s conclusion, we share some thoughts on the capital formation process of the hedge fund
industry, alpha capacity, and a research agenda much in need of input.

Recent Growth in the Hedge Fund Industry
We start with an overview of recent developments in the hedge fund industry—size,
number of funds, style composition, and fees.
Increasing institutional demand for hedge funds. Demand for hedge funds
by U.S. institutional investors has been steadily growing in the past five years.
According to the National Association of College and University Business Officers
(NACUBO) Annual Endowment Survey, the dollar amount and percentage of assets
invested in hedge funds have been steadily rising. As shown in panel A of Table 1, on an
equally weighted basis, endowments have increased allocation from 3.1 percent in
1999 to 8.7 percent in 2005. On a dollar-weighted basis, the increase is even more
dramatic, from 5.1 percent in 1999 to 16.6 percent in 2005. This growth has taken
place across the board, for small as well as large endowments. The dollar amounts in
panel B show an increase of $11.3 billion in 2000 to $49.6 billion in 2005.
Following the lead of university endowments, U.S. pension plans are also increasing their investments in hedge funds. According to Pensions and Investments, the
largest 200 U.S. defined-benefit pension plans invested $3.2 billion, or 0.1 percent of
their assets, in hedge funds in 2000. This investment grew to $29.9 billion, or 0.8 percent of their assets, in 2005.
Growth in number of funds and assets under management. The supply of
hedge funds has grown with the increase in demand for hedge funds, but the actual
size of the hedge fund industry is harder to measure. Unlike the mutual fund industry, hedge funds do not have an industry association to collect and report their information. Instead, hedge funds voluntarily provide information to one or more database

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Table 1
U.S. Institutional Assets Allocated to Hedge Funds
All university
endowment
(equally weighted)

All university
endowment
(dollar-weighted)

Top 200
DB pension
(dollar-weighted)

A: Percent of assets
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005

0.7
1.5
1.6
1.8
2.2
2.8
3.1
3.0
4.2
5.1
6.1
7.3
8.7

5.1
4.7
6.1
11.3
13.5
14.7
16.6

0.11
0.33
0.50
0.66
0.84

B: Amounts (in $billions)
2000
2001
2002
2003
2004
2005

11.3
14.4
25.1
31.1
39.2
49.6

3.2
8.5
14.4
21.1
29.9

Source: NACUBO, Annual Endowment Survey, various issues; Pensions and Investments, various issues

vendors. The lack of a uniform reporting standard makes it difficult to assess the true
size of the hedge fund industry.
There are now three commercial databases of hedge funds each having more than
ten years of actual data collection experience: the Center for International Securities
and Derivatives Markets (CISDM) at the University of Massachusetts in Amherst,
Hedge Fund Research (HFR) in Chicago, and Lipper TASS (TASS).2 In this section,
we exclude funds of funds (FoFs) from consideration to avoid double counting, since
FoFs invest in other hedge funds. As of December 2004, TASS had 4,130 funds (2,431
live and 1,699 defunct), HFR had 5,158 funds (2,939 live and 2,219 defunct), and
CISDM had 3,246 funds (1,315 live and 1,931 defunct).
While many hedge funds report to only a single database, some hedge funds
report to more than one database. An ongoing project at the BNP Paribas Hedge
Fund Centre of the London Business School is merging several databases to achieve
2. There are three notable entrants to this field—Morgan Stanley Capital International (MSCI),
Eureka Hedge, and Standard and Poors (S&P). Because of their late entry to this field, their data
were largely from reconstructed history rather than real-time collection of hedge fund performance. In this paper, we use the TASS database as of February 2005, HFR as of January 2005, and
CISDM as of December 2004. We report the results up through December 2004.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Figure 1
The Hedge Fund Universe in 2005: TASS, HFR, CISDM, Eureka Hedge, and MSCI

CISDM

20%

2%
2%

TASS

EUR

<1%
<1%

15%

6%
1%

<1%
3%

2%
<1%

<1%

<1%
<1%

<1%

1%
1%

<1%
1%

16%
7%
HFR
MSCI

Source: CISDM, Eureka Hedge, HFR, MSCI, TASS

a comprehensive picture. A great deal of effort has been expended to obtain an accurate assessment of the statistical characteristics of the hedge fund industry, eliminating the risk of double counting due to the lack of a uniform reporting standard in
the industry.3 At the completion of this project, we will have one of the most comprehensive academic hedge fund databases to work with. An early output of this project is Figure 1, which reports the differences among five databases in the form of a
Venn diagram.
At the present stage, we quantify the growth of the industry using the three
databases, shown in Table 2. In TASS, the number of hedge funds grew from 1,778 at
the end of 1999 to 2,431 at the end of 2004. The growth rate of 37 percent resulted from
2,111 new funds and 1,258 exiting funds. In HFR, there were 2,062 funds at the end of
1999, growing by 43 percent to 2,939 at the end of 2004, with 2,552 entries and 1,478
exits. In contrast, the number of funds in CISDM actually fell from 1,470 to 1,315
between 1999 and 2004, with 1,372 entries and 1,412 exits.
A comparable picture emerges from the estimates of industry totals made by various consultants. For example, the HFR Industry Report for the third quarter of 2005
estimated that the number of funds grew from 3,102 in 1999 to 5,782 in 2004 and that
AUM grew from $456 billion in 1999 to $973 billion in 2004.
From our earlier thesis of the hedge fund business model, a hedge fund is more
akin to a start-up than to a mutual fund. The high attrition rates (around 10 percent

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Table 2
Number of Funds in TASS, HFR, and CISDM
TASS (as of February 2005)

HFR (as of January 2005)

CISDM (as of December 2004)

Year

Start

Entry

Exit

End

Start

Entry

Exit

End

Start

Entry

Exit

End

1994

650

211

832

823

270

1,070

513

224

712

1995

832

249

1,023

1,070

384

1,399

712

220

834

1996

1,023

295

119

1,199

1,399

374

171

1,602

834

293

1,040

1997

1,199

316

1,425

1,602

374

172

1,804

1,040

334

117

1,257

1998

1,425

307

145

1,587

1,804

381

320

1,865

1,257

290

192

1,355

1999

1,587

367

176

1,778

1,865

419

222

2,062

1,355

313

198

1,470

2000

1,778

365

206

1,937

2,062

407

310

2,159

1,470

256

191

1,535

2001

1,937

384

229

2,092

2,159

451

251

2,359

1,535

249

216

1,568

2002

2,092

406

233

2,265

2,359

489

252

2,596

1,568

257

233

1,592

2003

2,265

339

221

2,383

2,596

457

245

2,808

1,592

164

234

1,522

2004

2,383

250

193

2,431

2,808

329

198

2,939

1,522

133

340

1,315

Source: TASS, HFR, CISDM

per year) in hedge funds are comparable to those of young firms. Hedge fund returns
also contain substantial idiosyncratic risk, typical of small undiversified firms.
Figure 2 illustrates the low correlation of hedge fund returns (versus the high correlation of mutual fund returns) to standard asset classes, updating the results
reported in Fung and Hsieh (1997). Here we use 2,082 hedge funds in TASS and
14,927 mutual funds from the Morningstar January 2005 CD. Each fund is required
to have at least thirty-six monthly returns but only the last sixty monthly returns if
the fund has a longer history. Each fund’s returns are regressed on eight asset classes
comparable to those in Fung and Hsieh (1997),4 and the distribution of the R 2s for
each group is tabulated. As before, hedge funds have much lower correlation with
the asset classes than mutual funds.
Changes in styles and strategies. Beyond having low correlation to standard
asset classes, hedge funds form a heterogeneous group that use many different strategies delivering returns, which can be quite different from each other.5 Consultants
classify hedge funds according to qualitative self-described styles. For example, TASS
classifies hedge funds into ten styles. Convertible arbitrage funds typically attempt
to extract value by purchasing convertible securities while hedging the equity, credit,
and interest rate exposures with short positions of the equity of the issuing firm and
other appropriate fixed-income related derivatives.. Dedicated shorts are funds that
specialize in short selling securities that are perceived to be overpriced, typically
equities. Emerging market funds specialize in trading the securities of developing
economies. Equity market neutral funds typically trade long-short portfolios of
3. We thank MSCI and Eureka Hedge for allowing us access to their databases.
4. These classes are MSCI North American equities, MSCI non-U.S. equities, IFC emerging market
equities, JPMorgan U.S. government bonds, JPMorgan non-U.S. government bonds, gold (London
a.m. fixing), the Federal Reserve trade-weighted dollar index, and the one-month eurodollar deposit
rate (previous month).
5. Fung and Hsieh (1997) found that the first five principal components explained less than 50 percent of the cross-sectional variation in hedge fund returns.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Figure 2
Distribution of R 2 versus Eight Asset Classes
20
Hedge fund
Mutual fund

Frequency (percent)

0
0

100

Distribution (percent)

Source: CISDM, HFR, TASS, Morningstar, Datastream

equities with little directional exposure to the stock market. Event driven funds
specialize in trading corporate events, such as merger transactions or corporate
restructuring. Fixed-income arbitrage funds typically trade long-short portfolios
of bonds. Macro funds bet on directional movements in stocks, bonds, foreign
exchange rates, and commodity prices.. Long/short equity funds are typically
exposed to a long-short portfolio of equities with a long bias. Managed futures
funds are specialists in futures trading, typically employing trend-following strategies. All other strategies are grouped into others. In a later section we will describe
the risk factors in many of these styles.
Based on a study performed by TASS, Table 3 shows how the style composition
has changed over the years. Regarding the number of funds, panel A shows the style
composition has been quite stable since 1999. For assets under management, panel B
shows that there has been a slight decline in macro funds (from 15 percent in 1999
to 10 percent in 2004) and long/short equity funds (from 45 percent to 32 percent)
with an increase in others (from 0.4 percent to 10 percent).
The HFR Industry Report done in September 2005 provides some additional
useful information. In terms of fund age, 13 percent of hedge funds are less than one
year old, 18 percent are between one and two years old, 15 percent are between two
and three years, 21 percent are between three and five years, and 33 percent are
more than five years old. In terms of AUM, 21 percent of hedge funds have less than
$10 million, 17 percent between $10 million and $25 million, 31 percent between $25 million and $100 million, 12 percent between $100 million and $200 million, and 19 percent have more than $200 million.
Management fees and performance fees. Current hedge fund fees are roughly
the same as they were in 1999. Table 4 shows the distribution of hedge fund fees.
Similar to mutual funds, hedge funds charge a fixed management fee as a percent of
net assets under management. Panel A in Table 4 shows that more than 70 percent
of hedge funds charge a management fee between 1 percent and 2 percent. However,

ECONOMIC REVIEW

Fourth Quarter 2006

Table 3
Changes in Style Composition of Hedge Funds

Year

Convertible
arbitrage (%)

Dedicated
shorts (%)

Emerging
market (%)

Equity
market
neutral (%)

FixedEvent
income
driven (%) arbitrage (%)

Macro (%)

Long/short
equity (%)

Managed
futures (%)

Other (%)

Total
number/
$billion

A: Number of funds
1994

832

1995

1,023

1996

1,199

1997

1,425

1998

1,587

1999

1,778

2000

1,937

2001

2,092

2002

2,265

2003

2,383

2004

2,431

B: Assets under management

Fourth Quarter 2006

0.3

57.0

1995

0.2

72.3

1996

0.3

0.5

99.1

1997

0.2

0.4

144.6

1998

0.4

153.8

1999

0.3

0.4

197.2

2000

0.3

217.7

2001

0.3

0.5

261.4

2002

0.5

4.5

310.3

2003

0.2

8.5

489.5

2004

0.2

10.1

673.8

Source: TASS

F E D E R A L R E S E R V E B A N K O F AT L A N TA

ECONOMIC REVIEW

1994

F E D E R A L R E S E R V E B A N K O F AT L A N TA

unlike mutual funds, hedge funds also
charge a performance fee. Panel B shows
that roughly 80 percent of them charge a
20 percent incentive fee.
TASS
HFR
CISDM
Style evolution and changing
(percent) (percent) (percent)
investor clientele. There has been a
A: Management fees
growing trend for hedge funds to evolve
0 percent–0.99 percent
5
3
4
away from single-strategy specialists into
1 percent–1.99 percent
73
72
78
multistrategy entities. A case in point is
2 percent–2.99 percent
20
22
17
the creation of a multistrategy category
in the CSFB/Tremont index in 2003. One
3 percent and up
3
2
1
might consider “multistrategy” to be a
B: Incentive fees
modern variation of traditional macro
0 percent–19.99 percent
10
10
18
funds. While macro funds are known for
20 percent
85
86
78
taking often highly leveraged directional
20.01 percent and up
5
4
4
bets on conventional asset classes such as
Source: TASS, HFR, CISDM
stocks, bonds, and currencies globally in
an opportunistic manner, nowadays multistrategy hedge funds allocate risk capital opportunistically among different hedge
fund strategies applied to global markets. Both macro and multistrategy hedge funds
can be thought of as tactical asset allocators of risk capital.
The emerging dominance of multistrategy hedge funds is consistent with our
thesis on the hedge fund business model. Successful hedge fund firms naturally grow
and diversify so as to dampen the impact of economic cycles on their performance.
Their growth and diversification are natural consequences of the desire to maximize
the enterprise value of the hedge fund management firm. The conventional qualitative approach to assessing the risk of multistrategy hedge funds is unlikely to yield
much insight. A risk-factor approach may well be the only alternative to describing
the performance characteristics of this important class of hedge funds.
On the investor side, consistent with our conjecture of an emerging clientele effect,
empirical evidence shows behavioral differences in the way capital is allocated between
investors seeking alpha versus investors seeking factor bets. Fung et al. (2006) find
that alpha investors have a steady flow of investments to hedge funds, while beta
investors exhibit return-chasing behavior similar to mutual fund investors.

Table 4
Distribution of Management Fees and
Incentive Fees in Live Funds

Statistical Issues in Hedge Fund Data
A proper study of hedge fund returns requires accurately measured data. Hedge fund
researchers have been aware of potential biases in hedge fund databases resulting
from the nature of the data collection process. Ackermann, McEnally, and Ravenscraft
(1999) pointed out that hedge fund databases have survivorship bias, liquidation
bias, backfill bias, and selection bias.6 Liang (2000) and Fung and Hsieh (2000b) provided additional discussions. In this section, we provide updates to these issues using
three commercial data sets—TASS, HFR, and CISDM.7
Selection bias. Since inclusion in a database is at the discretion of a hedge fund
manager, hedge fund databases can suffer from selection bias. This bias arises when
the hedge funds in a database are not representative of the universe of hedge funds.
It is difficult to estimate the selection bias since we are not able to observe funds that
are not part of a database.
It is tempting to theorize the direction of selection bias. Since hedge funds are not
allowed to advertise, the only way to gain access to investors is to participate in a

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

hedge fund database. One might expect
Table 5
Average Annual Returns, 1994–2004
funds that have superior performance
would enter a database to attract investors.
TASS
HFR
CISDM
However, there is a counterbalancing
(percent) (percent) (percent)
argument. Many successful funds that are
Live
14.4
14.3
15.5
closed to new investors choose not to be a
Live + defunct
12.0
12.5
13.1
part of a database. Thus, neither the magLive + defunct
nitude nor the direction of the net effect
(excluding first 14 months) 10.5
11.2
11.6
of selection bias is clear, whether the
Survivorship bias
2.4
1.8
2.4
funds in a database have higher or lower
Incubation bias
1.5
1.4
1.5
returns than funds not in a database. In
Source: TASS, HFR, CISDM
the future, the selection bias effect could
be estimated if we could gain access to
private databases that include hedge funds not in a publicly available database.
Survivorship bias. A natural consequence of our hedge fund business model is
the simple prediction that a hedge fund ceases to be a viable business proposition if
the fund cannot achieve the economy of scale to provide the fund manager adequate
operating leverage. For operational funds, this fateful endpoint usually happens
when investors are disappointed by the fund’s performance and vote with their feet
by redeeming their capital. It is also possible that the fund-raising effort of the hedge
fund manager failed to attract the critical mass necessary to make the fund a viable
business proposition. Without reference to the outlier of undiscovered jewels, it is
generally true that surviving (or live) funds have better returns than dead funds. This
point has been long recognized in the mutual fund literature, for example, Brown et al.
(1992) and Malkiel (1995). Following that literature, we measure survivorship bias as
the average return of surviving funds in excess of the average return of all funds, both
surviving and defunct.
Table 5 provides the annualized average return of live hedge funds in the three
databases from 1994 until 2004. It is 14.4 percent for TASS (14.3 percent for HFR and
15.5 percent for CISDM). The average return of “live + defunct” funds is 12.0 percent
for TASS (12.5 percent for HFR, 13.1 percent for CISDM). Therefore, the survivorship bias is 2.4 percent for TASS (1.8 percent for HFR, 2.4 percent for CISDM). These
percentages are consistent with the estimates in prior research using earlier samples,
and they are larger than the survivorship bias in mutual funds, typically found to be
between 0.5 percent and 1.5 percent.
As the industry matures, we believe the severity of survivorship bias will be
reduced, based on the following insight. Hedge funds become defunct for various
reasons. Poorly performing funds either liquidate or stop reporting their returns,
causing survivorship bias. However, successful funds that are closed to new investments often stop reporting to databases. This latter type of defunct fund would not
create a survivorship bias. As of December 2004, among defunct funds in HFR, 41 percent were liquidated, 13 percent were closed to new investments or no longer reporting, and the remaining 46 percent were not reporting. The liquidated defunct group
6. Fung and Hsieh (1997) used a database of surviving funds. Unable to collect information on delisted
funds, they explicitly acknowledged their shortcomings. Their analysis of the styles and risk factors
of hedge funds should not be significantly affected by this data issue.
7. To avoid introducing noise from currency fluctuations, only funds that report in U.S. dollars are
included in this update. In general, very few hedge funds do not have a version in U.S. dollars.
Therefore, focusing only on U.S. dollar–denominated funds helps avoid errors that may arise from
duplicated funds that are quoted in different currencies.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

had an average annualized return of 7.2 percent during the 1994–2004 period while the
other two groups both had average returns
Fund age
TASS
HFR
CISDM
ALL
of 10.8 percent. As the industry matures,
in months (percent) (percent) (percent) (percent)
the proportion of the two latter types of
1
33
89
69
85
defunct funds is likely to increase, mitigat2
10
37
71
36
ing the severity of survivorship bias.
3
22
43
44
40
Incubation bias (backfill or instant
4
40
39
71
48
history bias). A new hedge fund usually
5
34
50
76
49
undergoes an incubation period to com6
28
31
43
34
pile a track record. If the track record is
7
50
41
61
48
respectable, the manager typically enters
8
30
25
67
36
the fund into a database, which is one of
9
48
46
50
48
the ways to attract the attention of poten10
29
53
85
51
tial investors. When a new fund enters the
11
36
47
61
46
database, its incubation history prior to
12
54
38
55
46
the entry date is “backfilled.” Thus, it is
13
57
36
81
54
natural to expect the early part of a fund’s
14
40
61
85
59
history to be biased upward—where no
15
61
39
72
54
lemons are on sale, at least during the
16
59
48
74
60
“honeymoon period.”
17
51
51
77
56
Since a fund does not disclose the
18
47
40
79
52
length
of its incubation period, we can
19
55
56
62
57
apply
our
simple hedge fund business
20
49
43
77
53
model
to
infer
what would be the reason21
52
51
77
58
able
length
of
an
incubation period. Let us
22
49
45
86
54
assume
that
the
opportunity
cost to a hedge
23
32
27
66
40
fund
manager
running
his
or
her own fund
24
57
30
52
44
is
simply
to
stay
as
an
employee
in a com25
38
47
78
51
parable
investing
institution,
for
example,
26
37
49
70
48
at
a
proprietary
trading
desk
of
an
invest27
63
43
60
53
ment
bank
or
in
another
asset
management
28
54
46
74
55
firm. Departing from the comfort of an insti29
39
59
75
55
30
55
36
51
47
tutional setting offers the challenge of
31
52
42
59
50
building a business with potential enter32
48
45
72
54
prise value. Also, it offers the hedge fund
33
57
50
68
57
manager the opportunity to diversify his or
34
44
61
74
57
her client base. Note that being employed
35
41
41
66
47
by a single institution is somewhat equiva36
63
30
31
38
lent to having only a single investor in one’s
60
45
47
64
51
fund. Against these benefits are the oppor120
37
40
53
43
tunity costs—a steady stream of income
>120
24
32
43
32
and the supply of working capital for the
Source: TASS, HFR, CISDM
business infrastructure.
These are the considerations that a
hedge fund manager must weigh during the incubation period. The opportunity costs
involved during a fund’s incubation period are likely to be substantial since most
managers of a new, fledgling hedge fund are likely to have worked in the lucrative
financial industry. Faced with these costs, most managers, logically, would expect the
incubation period not to exceed a couple of years on average.

Table 6
Dropout Rate by Fund Age

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

This economic inference is consistent with the dropout rate of hedge funds from
databases. Table 6 provides the hazard rate, that is, the fraction of funds dropping out
of a database at a given age, averaged over TASS, HFR, and CISDM. The highest dropout
rate occurs when a fund is fourteen months old. Of course, in order to drop out of a
database, a fund must have entered it first. Thus, the fund age with the highest dropout
rate is a reasonable estimate of the length of the incubation period. Incidentally, the high
dropout rate in hedge funds is similar to
The high attrition rates in hedge funds
that in venture capital firms, in which typically only a low percentage (roughly 10 perare comparable to those of young firms.
cent) of new firms succeed.
Hedge fund returns also contain substanUsing the estimate of fourteen months
tial idiosyncratic risk, typical of small
as the incubation period, we deleted the
first fourteen months of each fund’s return.
undiversified firms.
As shown in Table 5, the average annual
return for “live + defunct” funds now drops to 10.5 percent in TASS, leading to the
estimate of 1.5 percent to be the incubation, or backfill bias. HFR and CISDM yield
virtually the same estimate (1.4 percent for HFR, 1.5 percent for CISDM).8 This result
is in line with previous research.
Some researchers use the information from TASS on the entry date of each
fund to estimate incubation bias. They treat all the data prior to entry as biased.
Unfortunately, this method can lead to extremely long incubation periods. This happens when a fund switches databases. For example, fund A stops reporting to HFR
and starts reporting to TASS. The part of its history after entry into HFR but before
entry into TASS is not backfill bias. A similar situation occurred after Tremont Capital
Management acquired TASS. Tremont added a significant number of funds from its
own database into TASS in September 2001. Since these funds were already tracked
by Tremont (but not by TASS), that part of their history subsequent to entry to
Tremont but prior to entry to TASS should not be treated as backfilled.
Finally, as the industry grows, the rise in demand for quantitative information on
funds in an electronic form is inevitable. More and more, hedge funds that were skeptical about the usefulness of hedge fund databases are slowly but surely overcoming
their aversion to reporting their performance. This trend underscores the importance
of understanding the economics of backfill bias versus the quirkiness of changes in
data collection methods.
Together, survivorship bias (roughly 2.5 percent) and incubation bias (around
1.5 percent) sum to roughly 4 percent per year. It is important to correct for these
biases, particularly when we study hedge fund excess returns (alpha) beyond exposure to systematic risk factors. As we shall see later on, hedge fund alphas are estimated to be around the same order of magnitude.
Liquidation bias. Hedge fund data suffer from an additional bias, called liquidation bias, that does not have a counterpart in mutual fund data. Liquidation bias
refers to the fact that hedge fund managers stop reporting returns to a database
before the final liquidation value of a fund. For example, several funds lost all of their
capital during the Russian debt crisis in August 1998. However, the managers did not
report returns of –100 percent in August 1998. Instead, the returns ended in July
1998. This practice causes an upward bias in the observed returns of defunct funds.
8. We also tried using other estimates of the incubation period: ten, fifteen, sixteen, twenty-two, and
twenty-seven months, based on different local peaks of the dropout rate in the three databases.
The resulting estimates of the incubation bias range from 1.1 percent to 1.9 percent.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

It also causes value-at-risk (VaR) models, based on observed fund returns, to underestimate the risk of hedge funds.
To estimate liquidation bias, we must be able to follow funds until liquidation.
Doing so, unfortunately, requires substantial resources. Instead, some researchers
make arbitrary assumptions about the return of a hedge fund in the liquidating
month, for example, –100 percent, as in
Posthuma and van der Sluis (2003). Such
Beyond having low correlation to standard
an approach seems extreme.
asset classes, hedge funds form a heterogeAckermann, McEnally, and Ravenscraft
neous group that use many different strate(1999) actually had estimates of liquidation
bias. They asked the data vendor (HFR in
gies delivering returns.
their case) to determine the liquidation
value of hedge funds, and they report the liquidating return of funds to be 0.7 percent (1999, 867–68). This return is certainly very far from the extreme assumption
of –100 percent.
In the future, it would be useful to settle this issue by asking data vendors to
determine the liquidation value of hedge funds. Another useful avenue to employ is to
directly approach the hedge fund administrators for more accurate records of the
final liquidation values of funds that ceased to operate.
Serial correlation of hedge fund returns. Krail, Asness, and Liew (2001)
observed that hedge fund indexes have serial correlation and that their returns are
correlated to past returns of market factors such as the S&P 500. This correlation can
be the result of infrequent trading of illiquid securities in their portfolios or of manipulation by managers to smooth their returns. Getmansky, Lo, and Makarov (2004)
provided a formal statistical model applied to individual hedge funds. Unfortunately,
neither Krail, Asness, and Liew (2001) nor Getmansky, Lo, and Makarov (2004) were
able to distinguish between the two causes of serial correlation in hedge funds—illiquid securities or return smoothing.
Over the past few years, administrative, accounting, and auditing service providers
to the hedge fund industry have been consolidating. Due diligence standards are
much higher with the ever-increasing presence of institutional investors in the hedge
fund industry. Gone are the days in which a small hedge fund partnership’s accounts
come from the manager’s accountant who happens to be his neighbor. In addition,
lenders to hedge funds such as prime brokers and investment banks also impose certain organizational standards on their hedge fund counterparties. Last but not least,
regulators are imposing compliance requirements on hedge fund companies that
operate in major capital markets. Taking all these factors together, we would argue
that, increasingly, hedge fund managers will find it difficult to manipulate the pricing
of their portfolios to smooth returns. Rather, illiquidity of the underlying market in
which a hedge fund transacts will more likely be the explanation.
The empirical evidence is certainly consistent with this line of reasoning. First,
we observe that hedge funds that trade liquid securities have returns that exhibit
little serial correlation. For example, the CSFB/Tremont Managed Futures Index
has a first-order autocorrelation of 0.07, which is not statistically different from
zero. This result is to be expected since managed futures funds tend to trade highly
liquid instruments and rely heavily on the liquidity of their position to achieve a
higher degree of leverage. In contrast, the CSFB/Tremont Convertible Arbitrage
Index has a first-order autocorrelation of 0.56, which is statistically different from
zero. Again, this outcome is hardly surprising since convertible arbitrage funds
tend to hold convertible bonds (hedging the credit risk with short positions in the

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

equity of the issuing firms), which transact primarily in over-the-counter markets.
One could go further to assert that it is precisely the provision of liquidity by convertible arbitrageurs that earns them the economic rent (see, for example, Agarwal
et al. 2006).
Second, we noted from our business model for hedge funds that prime brokers
(and investment banks) are the main suppliers of leverage to hedge funds. Given
the normal conflict between lenders and borrowers regarding position-risk sharing,
it is highly unlikely that prime brokers would share the same optimistic valuation
that would allow hedge fund managers to manipulate asset prices to smooth
returns. More often than not, the VaR models used by prime brokers (lenders) tend
to err on the conservative side when pricing portfolio positions. Since hedge fund
auditors naturally have access to prime broker reports, it is not likely that the auditors would allow hedge fund managers to use their own prices that are materially
different from the prime brokers’ to determine the value of the fund. Nonetheless,
more research is needed to determine which of these competing alternatives—illiquidity or return smoothing—is the culprit for the observed serial correlation of
hedge fund returns.
Implication for hedge fund benchmarks. The biases in hedge fund databases
cast doubt on the extent to which hedge fund indexes derived from these databases
accurately reflect actual investment experience. The potential for performance
divergence between an index of performance statistics and its executable clone is
particularly evident in the performance of investable indexes. Circa April 2003, a
number of passive portfolio products came to market intending to mimic the return
characteristics of published indexes of hedge funds. Despite promising prelaunch pro
forma returns, these investable indexes have substantially underperformed their
respective style benchmarks. For example, the HFRX Equity Hedge Investable Index
has an annualized return of 6.6 percent from its inception in April 2003 until
September 2005, which is less than half of the 14.1 percent annualized return of the
corresponding HFRI Equity Hedge Index over the same period.
The reasons for such a dramatic divergence in performance are many and varied.
They range from the effect of transaction costs associated with managing the mimicking portfolio to the unrealistic assumption of unlimited access to often scarce
capacity of some hedge funds—an inherent assumption that is present in most indexes
of hedge fund performance, albeit at varying degrees depending on the index construction method. More general issues of hedge fund benchmarks are discussed in
the following section. Suffice it to say that the gulf between an index of performance
statistics (hedge fund index) and practice (actual investment experience of hedge
fund investors) can be considerable.
An alternative is to use the average return of funds of funds (FoFs), as suggested in Fung and Hsieh (2000b). FoFs are actual pools of hedge funds, and, as
such, they directly reflect actual investment experience in hedge funds. The
databases on FoFs are less prone to biases (such as survivorship, incubation, etc.)
than those on individual hedge funds. Finally, the net (after FoFs fees) performance of FoFs is the net of the costs (due diligence and portfolio construction) of
investing in hedge funds borne by the investors, which are typically not reflected
in the returns of individual hedge funds.9 More discussions on this and related topics can be found in a later section.
9. While the investable indexes are FoFs, we are unable to determine the amount of fees charged
in these products.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Research on Hedge Fund Risk Factors
In this section, we address the fundamental question posted in the introduction: Are
there systematic risk factors inherent in hedge fund returns? It is important to note
that the data biases discussed in the previous section generally do not affect the
types of risk factors we find in hedge funds. Rather, data biases tend to distort estimates of hedge fund alphas.
Much research effort has been devoted to understanding the risk of hedge funds,
particularly in relating hedge fund returns to observable market risk factors. In general, research to date can be characterized as bottom-up versus top-down. Given the
heterogeneity of hedge funds, it is natural to start from a bottom-up approach, where
a number of papers have focused on identifying the risk factors inherent in specific
styles. For example, Fung and Hsieh (2001) linked the return of managed futures
(trend followers) to option straddles. Mitchell and Pulvino (2001) tracked the returns
of merger arbitrage funds to a passive merger arbitrage strategy. Duarte, Longstaff,
and Yu (2005) studied fixed-income arbitrage strategies replicating the returns of
commonly used strategies based on observable prices of fixed-income securities and
their derivatives. Agarwal et al. (2006) replicated convertible arbitrage (CA) returns
by mimicking a family of CA strategies using data from the underlying securities. The
subsections that follow provide similar analysis of the nine major hedge fund styles
described earlier.
From a top-down perspective, the question can be phrased as follows: In a diversified portfolio of hedge funds, what are the irreducible risk factors? This question
has been tackled in Fung and Hsieh (2003, 2004b). Taking the two approaches
together, this section reviews the status of research starting with strategy-specific
work leading up to hedge fund portfolio factors.
Risk factors of managed futures (trend followers). The majority of managed
futures funds employ a trend-following strategy. Fung and Hsieh (2001) extended the
theoretical model in Merton (1981) from market timers to trend followers. Merton
(1981) showed that a market timer, who switches between stocks and Treasury bills,
generates a return profile similar to that of a call option on the market. Fung and Hsieh
(2001) generalized this insight on a market timer to that of a trend follower, who seeks
to profit from large up-and-down movements using both long and short positions. The
resulting return profile is similar to that of a lookback straddle.10
Using exchange-traded standard straddles in twenty-six markets, Fung and Hsieh
(2001) replicated returns of lookback straddles. They constructed five portfolios
of lookback straddles—stock indexes, bond futures, interest rate futures, currency
futures, and commodity futures. Empirical evidence shows that three option portfolios—bonds, currencies, and commodities—have strong correlation to trend followers’ returns at a level well beyond previous results. It is intuitively appealing that
market volatility has been a key determinant of trend-following returns. Since the
Fung-Hsieh (2001) study, the relationship they reported has continued to hold.
Figure 3 provides evidence on the usefulness of the lookback portfolios. The heavier line represents the monthly returns of trend followers (based on the CISDM Trend
Following Index). The out-of-sample return forecasts of trend followers are generated using the regression coefficients from the regression equation in Fung and Hsieh
(2001), which was fit to data ending in 1997, and the realized values of the lookback
portfolios starting in 1998. The forecast returns continued to track the actual returns
of trend followers after the Fung and Hsieh (2001) study.
Risk factors in merger arbitrage. Mitchell and Pulvino (2001) created an
index for merger arbitrage returns, using announced mergers from 1964 until 2000.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Figure 3
Actual and Predicted Returns of Trend Followers, 1998–2004
15

Monthly returns (percent)

Predicted

–5
Actual

–10

–15
1998

1999

2000

2001

2002

2003

2004

Source: CISDM, Chicago Board of Trade (CBOT), Chicago Mercantile Exchange (CME), London International Financial Futures Exchange (LIFFE),
New York Mercantile Exchange (NYMEX), Sydney Futures Exchange (SFE)

They showed that the merger arbitrage returns are similar to those of merger arbitrage hedge funds. In fact, they observe that both the merger arbitrage index and
merger arbitrage funds exhibit characteristics similar to a dynamically adjusted short
position on the stock market. These results are illustrated in Figure 4.
This observation is intuitively appealing. Essentially, merger arbitrageurs are betting on the consummation of a merger—in general, they are long “deal” risk. Their
return can be viewed as the insurance premium from selling a policy against the failure to complete a merger. Typically, mergers fail for idiosyncratic reasons and can be
diversified away in a portfolio of such transactions. However, when the stock market
has a severe decline, mergers tend to be called off for a variety of reasons—ranging
from funding and pricing issues to concerns over the long-term prospects of the
economy. This scenario is one in which there is a convergence of deal risk that cannot
be easily diversified.
We note here that merger arbitrage (also known as risk arbitrage) is a substrategy
in the event-driven style discussed earlier. The other substrategy is distressed securities, which is covered in a later section.
Risk factors in fixed-income hedge funds. Fung and Hsieh (2002) analyzed
fixed-income hedge funds. They showed that convertible bond funds were strongly correlated to the CSFB Convertible Bond Index. High-yield funds were strongly correlated
to the CSFB High-Yield Bond Index. In addition, all styles, including fixed-income arbitrage (one of the major hedge fund strategies listed earlier), have correlations to
changes in the default spread.
10. A lookback straddle consists of a lookback call option and a lookback put option. A lookback call
option allows the owner to buy the underlying asset at the lowest price during the life of the call
option. A lookback put option allows the owner to sell the underlying asset at the highest price
during the life of the put option. The lookback straddle therefore allows the owner to buy at the
low and sell at the high. The lookback option was analyzed by Goldman, Sosin, and Gatto (1979).

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Monthly excess return of merger aribtrage hedge funds (percent)

Figure 4
Risk Factor for Merger Arbitrage Hedge Funds, 1994–2004
3
2
1
0
–1
–2
–3
–4
–5
–6
–7
–20

–15

–10

–5

Monthly excess return of S&P 500 (percent)

Source: HFR, Datastream

Figure 5 provides support for this view. Here, we graph the HFR Fixed-Income
Index against the change in credit spread, as proxied by Moody’s Baa yield over the
ten-year Treasury constant maturity yield.
In a more recent study, Duarte, Longstaff, and Yu (2005) created returns using
various fixed-income arbitrage trades frequently used by hedge funds—swap spreads,
yield-curve spreads, mortgage spreads, fixed-income volatility arbitrage, and capital
structure arbitrage.
Essentially, the swap spread trade is a bet that the fixed side of the spread (the
difference between the swap rate and the yield of the Treasury security of the same
maturity) will remain higher than the floating side of the spread (the difference
between LIBOR and the repo rate) while staying within a reasonable range that can
be estimated from historical data. Yield-curve spread trades are “butterflies,” betting
that bond prices (which can be mapped to points along the yield curve) deviate from
the overall yield curve only for short-run, tactical liquidity reasons, which dissipate
over time. Mortgage spread trades are bets on prepayment rates, consisting of a long
position on a pool of GNMA mortgages financed using a “dollar roll,” delta-hedged
with a five-year interest rate swap. Fixed-income volatility trades are bets that the
implied volatility of interest rate caps tends to be higher than the realized volatility
of the eurodollar futures contract. Capital-structure arbitrage or credit arbitrage trades
on mispricing among different securities (for example, debt and equity) issued by the
same company.
Duarte, Longstaff, and Yu (2005) found strong correlation between the returns
of these strategies and the returns of fixed-income arbitrage hedge funds. In addition, many of these strategies have significant exposure to risks in the equity and
bond markets.
Risk factors in long/short equity hedge funds. As discussed earlier, the
long/short equity style accounts for 30 to 40 percent of the total number of hedge
funds. Agarwal and Naik (2004) studied equity-oriented hedge funds, and Fung and

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Figure 5
Risk Factor for Fixed-Income Hedge Funds, 1990–2004

Monthly excess return of fixed-income hedge funds (percent)

6
5
4
3
2
1
0
–1
–2
–3
–4
–5
–1

Monthly change of Moody’s Baa yield—ten-year Treasury yield (percent)

Source: HFR, Board of Governors of the Federal Reserve System (BOG)

Hsieh (2004a) focused on long/short equity funds. Basically, there is strong evidence
that long/short equity funds have positive exposure to the stock market as well as
exposure to long small-cap/short large-cap positions, similar to the SMB factor in the
Fama-French (1992) three-factor model for stocks.
Figure 6 provides support for this view. Here, we use the previous twenty-four
months of data to estimate the exposure of long/short equity funds (as proxied by the
HFRI Equity Hedge Index) to the S&P 500 and the difference between the Russell 2000
and S&P 500. The estimated coefficients are used to perform a one-month-ahead
conditional forecast.11 The figure shows that the one-month-ahead forecast is a very
good predictor of the HFRI Equity Hedge Index return. An intuitive explanation of
these results is as follows. Typically, long/short equity hedge fund managers are stock
pickers with diverse opinions and ability. As such, individual performance of these
managers is likely to be highly idiosyncratic. However, all managers are subject to the
basic phenomenon that “underpriced stocks,” if they exist, are likely to be found among
smaller, “under-researched” stocks. On the short side, liquidity condition in the stockloan market makes small stocks much less attractive candidates for short sales. There
is also the question as to whether long/short hedge fund managers exhibit market
timing ability—a topic for ongoing research. Thus far, the empirical evidence does not
lend support to such a proposition (see, for example, Fung and Hsieh 2006).
Risk factors in convertible arbitrage. Using U.S. and Japanese convertible
bonds, Agarwal et al. (2006) created returns for three basic strategies frequently
employed by convertible arbitrage funds. The volatility arbitrage strategy is a bet that
the embedded option in the convertible bond is mispriced. The credit arbitrage strategy
is a bet that the credit risk in the convertible bond is mispriced. The carry strategy is
a combination of these two strategies. The results point to convertible arbitrage hedge
11. Specifically, the one-month-ahead conditional forecasts use the regression coefficients from the
previous twenty-four months and the realized values of the regressors in the subsequent month.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Figure 6
Actual and Predicted Returns of Long/Short Equity Hedge Funds, 2003–04
5
4
Actual

Monthly returns (percent)

3
2
1
Predicted

0
–1
–2
–3
Jan.
2003

April
2003

July
2003

Oct.
2003

Jan.
2004

April
2004

July
2004

Oct.
2004

1.5

Source: HFR, Datastream

Monthly excess return of convertible arbitrage hedge funds (percent)

Figure 7
Risk Factor for Convertible Arbitrage Hedge Funds, 1993–2002
4
3
2
1
0
–1
–2
–3
–4
–2

–1.5

–1

–.5

2.5

Fitted values using U.S. and Japanese convertible arbitrage strategies (percent)

Source: HFR, Agarwal et al (2006)

funds as providers of liquidity to the convertible bond market trading mainly from the
long side while hedging the inherent risk factors of the bond. One interpretation of
the Agarwal et al. (2006) results is that the return to convertible arbitrage hedge
funds stems from a liquidity premium paid by issuers of convertible bonds to the
hedge fund community.
The returns from these strategies can explain a significant part of the return variation in convertible arbitrage funds, which is illustrated in Figure 7. As in Agarwal et al.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Monthly excess return of emerging market hedge funds (percent)

Figure 8
Risk Factor for Emerging Market Hedge Funds, 1994–2004
20
15
10
5
0
–5
–10
–15
–20
–25
–30

–25

–20

–15

–10

–5

Monthly excess return of IFC Emerging Market Index (percent)

Source: HFR, Datastream

(2006), the excess return of the HFRI Convertible Arbitrage Index is regressed on
the excess returns of the Vanguard Convertible Securities Portfolio (as a proxy for
the underlying convertible bonds) and the U.S. and Japanese volatility arbitrage
strategy, credit arbitrage strategy, and carry strategy. The adjusted R 2 of the regression is 0.38. The observed returns are graphed on the vertical axis, while the fitted
values of the regression are graphed on the horizontal axis. The upward-sloping line
is evidence that the risk factors capture a significant part of the variation in the
returns of convertible arbitrage funds.
Risk factors in niche styles. This section summarizes the research findings on
risk factors inherent in the other hedge fund styles. Figure 8 shows that the hedge
fund returns of emerging market hedge funds are strongly correlated with the IFC
Emerging Market Stock Index.
Figure 9 shows that distressed securities hedge funds’ returns are strongly correlated with the CSFB High-Yield Bond Index. (As noted earlier, distressed securities is
one of the two substrategies in the event-driven style, along with merger arbitrage.)
HFR has an index called Equity Non-Hedge, consisting of hedge funds that typically trade from the long side, leaving their market risk largely unhedged. Figure 10
shows that their monthly excess returns are strongly correlated with the Wilshire
Small Growth Stock Index. Figure 11 shows that dedicated short-sellers’ returns are
strongly negatively correlated with the Wilshire Small Growth Stock Index.
Thus far, we have covered the risk factors in seven of the nine styles described
earlier (excluding others). Only two styles remain—macro and equity market neutral.
Macro funds are analyzed in the next section. Equity market neutral has been a problematic style to analyze, largely because of difficulties with accurate classification of
strategies that fall into this category. Many funds in HFR and TASS carrying the equity
market neutral classification exhibit statistically significant betas to standard equity
factors; see, for example, Patton (2004). In addition, there is another category in the
HFR database, called statistical arbitrage, comprising long/short equity hedge funds

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Monthly excess return of distressed securities hedge funds (percent)

Figure 9
Risk Factor for Distressed Securities Hedge Funds, 1994–2004
6
4
2
0
–2
–4
–6
–8
–10
–8

–6

–4

–2

Monthly excess return of CSFB high-yield bonds (percent)

Source: HFR, Morningstar

Monthly excess return of HFRI Equity Non-Hedge Funds (percent)

Figure 10
Risk Factor for Long-Biased Equity Hedge Funds, 1994–2004
15

–5

–10

–15
–30

–20

–10

Monthly excess return of Wilshire Small Growth Index (percent)

Source: HFR, Wilshire

that primarily employ quantitative models to construct factor-neutral portfolios.
Statistical arbitrageurs that operate in equity markets should naturally fall within the
equity market neutral category. It is unclear to us how these distinct categorizations
are made. Taken together, one observes that there are instances in which equity market neutral funds have significant betas to equity factors, whereas there are zero-beta
equity-related strategies that are being placed in a different category. More work is
required to arrive at an unambiguous definition of this hedge fund style.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Monthly excess return of HFRI Short-Selling Hedge Funds (percent)

Figure 11
Risk Factor for Short Selling Hedge Funds, 1994–2004
30
25
20
15
10
5
0
–5
–10
–15
–20
–25
–30

–20

–10

Monthly excess return of Wilshire Small Growth Index (percent)

Source: HFR, Wilshire

Capturing the risk in macro hedge funds. Macro fund managers are known
to be highly dynamic traders who often take highly leveraged bets on directional
movements in exchange rates, interest rates, commodities, and stock indexes.
Capturing the risk of macro funds has been a challenge to researchers given the
dynamic nature of the risk in these funds. Here, we present a simple, linear risk factor model that can capture a substantial amount of the risk to which macro funds are
exposed. It turns out that the seven-factor model of Fung and Hsieh (2004b)
(described in more detail below), which was designed for describing the inherent
risks in diversified portfolios of hedge funds, does a reasonable job in capturing the
risk of macro funds.
Figure 12 depicts the actual and one-month-ahead conditional forecast of the
HFRI Macro Index. As in the case of the forecasting exercise for long/short equity
funds in Figure 6, for each month we use the prior twenty-four months’ data to
regress the macro index on the seven risk factors. We use the regression coefficients
and the realized values of the seven risk factors in the subsequent month to generate the conditional one-month-ahead forecast.
Perhaps these results are not too surprising because the best-known macro funds
are large funds that tactically allocate risk capital across a portfolio of “bets” across
global markets. Each bet can be thought of as a directional strategy on a portfolio of
global securities—for example, holding long government bonds of one country while
funding the position with a short position in fixed-income instruments denominated
in another currency so as to maximize the upside potential of the long position at the
lowest cost of leverage globally available. The traditional view of a macro fund’s portfolio being guided by a very top-down view of global economic factors is, perhaps, not
too different from the tactical strategy allocation process of a portfolio of different
hedge fund strategies. Therefore, tools that work well in capturing the risk characteristics of diversified hedge fund portfolios can be applied in describing the risk in
macro funds.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Figure 12
Actual and Predicted Returns of HFRI Macro Funds, 1996–2004
8
Actual

Monthly returns (percent)

4
2
0
–2
–4
Predicted

–6
–8
1996

1997

1998

1999

2000

2001

2002

2003

2004

Source: HFR, BOG, Datastream, CBOT, CME, LIFFE, NYMEX, SFE

The challenge to academic researchers is to devise models that properly measure
the value these tactical strategy allocation decisions bring to investors.
A basic risk-factor model using asset-based style (ABS) factors. In this
section we discuss a top-down approach to modeling the risk factors of diversified
hedge fund portfolios. Understanding these risk factors is critical for investors, counterparties, and regulators alike. Investors need to identify the major risk factors in
hedge fund portfolios in order to assess the impact on their overall asset allocation
profile. Similarly, counterparties to hedge funds and their regulators need to understand the major sources of hedge fund risk to measure capital at risk.
Fung and Hsieh (2004b) proposed a basic risk-factor model using seven risk factors to account for the risk of diversified portfolios of hedge funds. These risk factors
are extracted from those that are found in empirical studies on many of the major
hedge fund styles. The excess return of the S&P 500 (SPMRF) and Small Cap minus
Large Cap (SCMLC) are the equity factors most important for long/short equity
funds, which comprise 30 to 40 percent of the entire industry. The return of the tenyear Treasury bond (BD10RET) above the risk-free return and the return of Baa
bonds above the return of the ten-year Treasury bond (BAAMTSY) are the bond factors most important for fixed-income hedge funds. The three lookback portfolios in
bonds (PTFSBD), currencies (PTFSFX), and commodities (PTFSCOM) are the key
risk factors for trend followers or managed futures.
Table 7 illustrates the efficacy of the seven risk factors in explaining the returns
of several standard hedge fund indexes—the HFRI composite index, the CSFB/Tremont
composite index, and the MSCI equally weighted composite index. In addition, the
seven risk factors can explain a high percentage of the variation in the HFR Fundsof-Funds Index.
Of general interest to investors is the estimate of a hedge fund’s alpha. For the
hedge fund indexes, alpha is between 11 and 27 basis points per month, or 1.32 percent to 3.24 percent per year on a net asset value (NAV) basis. As mentioned earlier,

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Table 7
Regression of Hedge Fund Composite Indexes on Seven Risk Factors,
April 2000–December 2004
Risk factors
Constant
SPMRF
SCMLC
BD10RET
BAAMTSY
PTFSBD
PTFSFX
PTFSCOM
R2

HFRI
composite

CTI
composite

MSCI
composite

HFR
FoF

0.0011
0.3126
0.2292
0.1661
0.1656
–0.0002
0.0108
0.0218
0.876

0.0017
0.1837
0.1650
0.2314
0.1103
–0.0043
0.0150
0.0186
0.612

0.0027
0.1653
0.1654
0.1794
0.0697
0.0020
0.0187
0.0083
0.763

–0.0006
0.1572
0.1408
0.1764
0.1941
–0.0021
0.0122
0.0184
0.691

Note: Bold figures indicate statistical significance at the 1 percent level.
Source: HFR, Tremont, MSCI, BOG, Datatream, CBOT, CME, LIFFE, NYMEX, SFE

this is on the same order of magnitude as survivorship bias (around 2.5 percent per
annum) and incubation bias (around 1.5 percent per annum). Interestingly, the average FoF does not have any positive alpha.
Note that these top-down risk factors are all based on traded securities and their
derivatives; hence we use the term asset-based to describe them. This term is appropriate as, by and large, hedge fund portfolios are composed of conventional securities and their derivatives. This recognition is important. Having identified readily
observable risk factors based on traded assets, we have indirectly circumvented the
opaqueness of hedge fund operations—at least for diversified portfolios. Here we
have a method for indirectly measuring the systematic risk of hedge fund investing
by observing market prices at higher frequency (than monthly NAVs of hedge funds)
and with much longer price history.
Figure 13 illustrates the usefulness of a long price history, in the form of the credit
spread (Moody’s Baa ten-year Treasury). Between the end of 1987 and September
1998 (the LTCM debacle), credit spreads had stayed in a narrow range relative to the
earlier years. This trend explains the particularly good performance and lack of large
losses of fixed-income hedge funds during that period, based on the evidence in
Figure 5 that showed fixed-income hedge funds as negatively exposed to this variable. As the credit spread widened after the Russian debt default in August 1998,
fixed-income hedge funds suffered large losses. While the increase in credit spreads
was large relative to the experience of the previous ten years, the spreads were not
especially large against the backdrop of a much longer time period. Therefore, it is
not surprising that LTCM, which by many accounts was many times more leveraged
than a typical fixed-income hedge fund, nearly failed.12
Risk monitoring and performance evaluation of hedge funds. The advantage of ABS factors is not only that their portfolio-level risk factors are much more
readily observable but that they are also investable. By construction, they are derived
from traded assets in the public market. This construction provides a much more
12. LTCM claimed that it experienced a “100-year” flood. But its risk management system reportedly
used only ten years of data, stopping around December 1987. Had it used ten more years of history, it might have avoided problems in 1998.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Figure 13
Long History of the Credit Spread

Moody’s Baa yield—ten-year Treasury yield (percent)

8
7
6
5
4

Dec. 1987
Sept. 1998

3
2
1
0
1925

1935

1945

1955

1965

1975

1985

1995

2005

Source: BOG

natural way of defining hedge fund alphas and hedge fund betas, which we refer to
as alternative alphas and alternative betas in Fung and Hsieh (2003).
The immediate application is clear. For investors, alpha buyers now have a way
to measure the quality of their hedge fund investment. Beta buyers (investors who
prefer leveraged factor bets) can assess whether their capital is exposed to the right
risk; and both can evaluate whether the fees they paid are appropriate. For counterparties, measuring the exposure to key risk factors offers a market-price-driven metric
that aggregates hedge fund risk in capital-at-risk calculations. For regulators, it provides a barometer to gauge potential convergence of systemic risk exposures from
hedge funds, proprietary desks, and conventional money managers. These issues are
explored further in the next two sections of the paper.

Hedge Funds and Regulatory Concerns
In the wake of the Asian currency crisis of 1997, the International Monetary Fund
(IMF) researched into complaints by Asian government officials that hedge funds
were the primary culprit for that episode. Their findings were published in May 1998
in Eichengreen et al. (1998). The paper addressed three primary regulatory concerns: investor protection, systemic risk, and market integrity. The authors noted
that “few regulators see a need for stricter regulation on the first two grounds” (1).
In terms of investor protection, regulators are generally of the view that investors
can “fend for themselves” (20). As private investment vehicles, hedge funds are available only to “accredited investors”—wealthy individuals and institutional investors.
These sophisticated investors have the savvy and the financial ability to complement
their knowledge by hiring consultants with the know-how to better understand the
risk in hedge funds. They also have sufficient wealth to withstand the risk of sizable
losses that might result from hedge fund investing.
In terms of systemic risk, regulators are concerned that banks under their supervision are exposed to counterparty risk in their transactions with hedge funds, but

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

“they regard these as problems best dealt with by existing supervision of banks and
other counterparties rather than by new regulations” (21).
On the issue of market integrity, Eichengreen et al. (1998) reported that there
were concerns “that hedge funds can dominate or manipulate particular markets”
(1). But they went on to say that “many observers are skeptical that hedge funds are
large enough to dominate markets.” Thus, they arrived at the conclusion that “the
case for supervisory and regulatory initiatives directed specifically at hedge funds is
not strong” (21).
In this section, we reflect on these arguments in light of the recent events and put forward a number of research questions searching for answers. Papers by Edwards (2006)
and Chan et al. (2006) at this conference cover these issues in much greater detail.13
Investor protection. Investor protection falls largely under the purview of the
Securities and Exchange Commission (SEC) and to a lesser extent the Commodity
Futures Trading Commission (CFTC). Up until 2003, the SEC has kept a fairly loose
rein on hedge funds, taking action usually only in fraud cases. In May 2003, the SEC
organized a Hedge Fund Roundtable to discuss “investor protection implications” of
hedge funds (see the SEC’s press release, 2003-40). A staff report (SEC 2003) was
published in September 2003, and the rule to require hedge fund advisers to register
as investment advisers was adopted in December 2004. SEC (2004) cited the growth
of the hedge fund industry, the increasing instances of hedge fund fraud,14 and the
broadening of exposure to hedge funds, especially by institutional investors (for
example, public and private pension funds, universities, endowments, foundations,
and charitable organizations) who had never before invested in hedge funds, as reasons for taking action to register hedge fund advisers.
While registration may not be a particularly onerous task for hedge funds, it is not
clear how it can effectively deal with the three concerns raised in SEC (2004).
In the first place, registration is not likely to slow down the proliferation of hedge
fund offerings to the small, unsophisticated investors that the SEC is concerned about.
It would be more effective to raise the requirement for “accredited investors” to make
it harder for less sophisticated investors to qualify for investing in hedge funds.
Second, hedge funds that accept money from pension plans are typically registered as investment advisers, a result of the ERISA Act of 1974.
Third, it is unclear how registration relates to the discovery and, ultimately, the
deterrence of hedge fund fraud. SEC (2004, 5) stated that fifty-one hedge fund fraud
cases were brought by the SEC for damages of $1.1 billion during 2000–04. It would
be helpful to have more research on how registration could have helped to prevent
fraud in these cases.
Systemic risk. Hedge funds can become the transmission mechanism of systemic
risk because they borrow from and trade with regulated financial institutions, such as
prime brokers and investment banks. Large losses from one or more hedge funds can
cause financial distress to the counterparties they deal with, which can in turn generate a domino effect to other financial institutions. This chain reaction is referred to as
systemic risk. The LTCM debacle put systemic risk front and center in the mind of regulators. The important point here is not to focus only on the risk of another LTCM but
13. Edwards (2006) can be found on page 35 in this issue of the Economic Review. An abbreviated
version of the Chan et al. (2006) paper presented at the conference is on page 49 of this issue.
14. SEC (2004, 5) cited fifty-one hedge fund fraud cases brought by the SEC during 2000–04, involving $1.1 billion in damages. The rule also referred to “almost 400 hedge funds (and at least 87 hedge
fund advisers)” that were being investigated at that time.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

to be aware of the risk of a convergence of opinion among different hedge funds on
the best trade(s). Fung, Hsieh, and Tsatsaronis (2000) described this as diversification implosion and provided empirical examples of this phenomenon.
Statistical analyses of hedge fund returns may furnish us with valuable lessons ex
post. However, these tools are generally incapable of sounding the alarm bell before
the fire engulfs the entire building. Ultimately, information about systemic risk exposures estimated from up-to-date position data is needed to offer any chance of providing an early warning. A substantial amount of this data is already in the hands of
hedge fund counterparties—prime brokers and prime bankers to hedge funds. The
question is how to consolidate the inforHedge funds can become the transmission mation into estimates of systemic risk exposures of the industry as a whole.
mechanism of systemic risk because they
Commercial service providers offering
borrow from and trade with regulated
consolidation of hedge fund position risk for
financial institutions, such as prime brolarge hedge fund portfolios are available
covering a significant subset of hedge funds
kers and investment banks.
that offer their products through managed
account platforms. This development is encouraging. Improved transparency by hedge
funds and the availability of better risk management tools to hedge fund investors
should help keep systemic risk in check. As things stand, investors are well placed to
rebalance hedge fund portfolios that are overly concentrated in a limited number of factor bets. To this end, investors’ interests and those of regulators are aligned. Both will
benefit from a continuing improvement in transparency from hedge fund managers.
However, a number of conflicts of interest that arise from the financing of hedge
funds by regulated financial institutions cannot be easily resolved. These can be
expressed using our simple hedge fund business model. In essence, the hedge fund
manager pledges the equity in the fund as collateral to financial intermediaries such
as prime brokers and investment banks against leveraged trading positions. As
lenders to the hedge fund, these counterparties hold the equivalent of senior debt
claims to the fund as a legal entity. Immediately, as lenders, they are exposed to the
downside risk of the hedge fund strategies they finance.
There is an obvious need to assess the systematic risk factors to which a hedge
fund strategy is exposed—in a similar manner to an investor of the fund. However,
the assessment of the upside potential is less critical (but not irrelevant) to lenders
compared to investors of a hedge fund. Good performance by a hedge fund potentially leads to more business for the prime brokers. In reverse, bad-performing funds
are not good for anyone—investors and prime brokers alike. However, what about
mediocre funds that generate transactional fees to the prime brokers but offer little
value to investors? Investors would certainly like to cut these funds from their portfolio, but would prime brokers be as quick to act since there is no apparent risk to
keeping these funds on their books? An interesting question is whether mediocre
funds searching for winning trades are more likely to herd—to become trend followers of popular trading ideas. The answer to this question will be important to
investors, financial intermediaries, and regulators.
Market integrity. A major concern of financial regulators is the potential impact
of hedge funds on financial markets. Hedge funds were mentioned in the Brady
Commission Report to be net buyers during the October 1987 stock market crash.15 In
an IMF staff report, Eichengreen et al. (1998) conducted interviews with market participants and hedge fund managers to gauge the role of hedge funds during various
market events, including the ERM (exchange rate mechanism) crisis in 1992, the

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Mexican peso crisis in 1994, and the Asian currency crisis in 1997. Generally speaking, Eichengreen et al. (1998) viewed hedge funds as too small to exert significant
impact on financial markets. Fung and Hsieh (2000a) provided some quantitative
support using hedge fund returns, including high-frequency weekly and daily data
collected from public sources.16
The view that hedge funds are too small to exert significant market impact was
challenged by the near bankruptcy of LTCM in 1998. The recent growth of the hedge
fund industry has also brought this issue back into the foreground for regulators. In
this section, we put forward the idea that the risk hedge funds pose to market integrity
has shifted from the likes of mega currency speculators or a highly leveraged powerhouse like LTCM to that of a convergence of leveraged opinions among funds that
individually may easily operate unnoticed. The avoidance of this phenomenon may
well require the action of better-informed portfolio investors—through greater transparency on the part of hedge fund managers and better access to risk management
tools—as well as more discriminating leverage providers to hedge funds, the financial intermediaries.
To this end, there are promising signs in the evolution path the hedge fund industry has taken. Transparency is improving, but more needs to be done. Risk management tools for hedge fund portfolios are becoming commonplace. Perhaps more
research and standardization in this area would help avoid a proliferation of risk management models that are as bewildering to consumers as opaque hedge funds. Capital
requirements for certain financial institutions—banks and insurance companies—
investing in hedge funds are directing them toward demanding more transparency
and risk management. Compliance requirements from regulators such as the Financial
Stability Authority (FSA) in London together with the institutionalization process of
hedge funds have greatly improved the operational integrity of hedge fund firms.
These are all helpful and healthy developments in the hedge fund market and certainly
look like a much more promising path than imposing direct, specific regulations on
hedge funds themselves.
Just as important has been the development in the supply side of hedge fund
products. Consistent with our simple model of the hedge fund business, the emergence of multistrategy hedge fund firms as a solution to the life cycles of hedge fund
strategies and the demand for institutional quality products have had a positive influence on the hedge fund industry. Hedge fund firms are becoming better organized,
better diversified, and more averse to erratic performance that could damage their
enterprise value. Perhaps we are witnessing the maturity of a hedge fund industry
that blends the best of the regulatory environment with rational economic behavior
on the part of investors, financial intermediaries, and hedge fund managers.

Alternative Alpha, Synthetic Hedge Funds, and the Price for Talent
In this section, we turn to the companion question of hedge fund systematic risk
(beta) frequently asked by investors: Do hedge funds achieve excess returns beyond
exposure to systematic risk factors?

15. The Brady Commission is properly known as the Presidential Task Force on Market Mechanisms
(1988).
16. More recently, Brunnermeier and Nagel (2004) investigated the role of hedge funds in the technology bubble of 1999–2000. They found little evidence that hedge funds were short tech stocks.
Instead, they found evidence consistent with the view that hedge funds rode the bubble on the
way up and then exited relatively quickly in early 2000.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

The search for alpha—have all the low-hanging fruits been picked? In our
opinion, the cleanest way to assess hedge fund alpha is through the returns of FoFs.
Data on FoFs have fewer biases than data on individual hedge funds, as pointed out
in Fung and Hsieh (2000b), and they reflect the actual investment experience in
hedge funds, netting out the cost of due diligence, portfolio construction, etc.
Fung et al. (2006) estimated the alpha of FoFs from the merged TASS, HFR, and
CISDM databases, using the seven risk factors from Fung and Hsieh (2004b). While
the average FoF does not deliver statistically significant alpha, about 22 percent have
positive alpha (significant at the 95 percent level). These “have-alpha” FoFs have a
higher probability of remaining “have-alpha” FoFs than the remaining “beta-only”
FoFs. They tend to have a steady inflow of capital, which does not exhibit returnchasing behavior. In contrast, the “beta-only” FoFs experience both inflows and outflows that have return-chasing behavior.
Additionally, Fung et al. (2006) found that alpha in both have-alpha and havebeta FoFs has declined in the recent period (April 2000 to December 2004) relative
to earlier periods. This decline coincides with the large inflow of money into the
hedge fund industry and is consistent with the prediction of Berk and Green (2004)
that fund flows will drive down the net-of-fee excess returns to zero so that there
should be no excess return to investors in equilibrium.
Can cheap hedge funds be created? If portfolios of hedge funds that are limited to ready-packaged hedge fund products are suffering from dwindling alphas, the
natural question that arises is, Can synthetic hedge funds be created at a lower cost
to investors?
There is a growing literature on replicating hedge fund returns using statistical
techniques. Here we provide a brief review of this approach. Generally, there are two
main issues to be resolved. It is natural to expect hedge fund managers to have
dynamic exposures to factor bets (beta bets). This conclusion naturally flows from
our simple hedge fund business model. Viewed as a business, the incentive is for the
hedge fund manager to maximize his or her enterprise value. To that end, the tendency is to diversify the income stream to the hedge fund management company.
This diversification process, we believe, has been the primary motive behind the
growth of multistrategy hedge funds (from 0.5 percent to 12–14 percent of total
capacity). As different strategies are introduced into the portfolio mix, different risk
factors will emerge and evolve over time. Statistical techniques without explicit
recognition of the changing risk factors are unlikely to be able to explain where the
next risk is coming from.
Replicating hedge fund alpha, on the other hand, takes us into the voluminous
literature on portable alphas. Fung and Hsieh (2004a) showed that there are
portable alphas in equity-related hedge fund strategies. However, we do not believe
that these alphas can be replicated by statistical means at a lower cost. After all, why
would anyone sell skill at below market-equilibrium price? Naturally, those who successfully generate persistent hedge fund alphas will simply join the ranks of hedge
fund managers and price their services accordingly. So long as alpha remains a
scarce commodity, new discoveries of alternative alpha sources are unlikely to
depress the market price for alpha.
Issues on compensating talent. Hedge fund returns are a mixture of alpha and
beta components with differing production costs and capacity constraints. This consideration leads us directly back to various incentive fee issues in the hedge fund
business model. Such a discussion necessarily involves the governance structure to
align managers’ incentives with those of investors. The paper by Lehmann (2006) at

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

this conference deals with these issues in greater detail. Here are some immediate
issues that flow directly from our discussions.
Presumably investors are more willing to pay fees for hard-to-replicate alphas
(and, to a lesser extent, hard-to-replicate factor bets or alternative beta factors) than
returns from more easily replicable conventional beta-like exposures. To date, very
few hedge fund contracts explicitly recognize these problems. For instance, rarely
The risk hedge funds pose to market integrity
does one come across risk-adjusted benchhas shifted to that of a convergence of levermarks being used as part of the incentive
aged opinions among funds that individufee hurdle. For now, there are no universally accepted investable performance
ally may easily operate unnoticed.
benchmarks for different hedge fund
strategies. The performance history of investable hedge fund indexes has thus far
raised more questions than answers to the whole question of benchmarking hedge
fund performance.
There is a companion problem to the benchmarking issue in designing incentive fee
contracts. Often, and in recognition of the deficiencies of the existing incentive fee contract, some investors have demanded co-investing—that is, that hedge fund managers
should invest a substantial amount of their personal wealth alongside the investors of
the funds they manage. This practice reduces the incentive for the manager to maximize the call option features embedded in the incentive contract by taking unjustifiable
risks; for a detailed discussion of these issues, see Goetzmann, Ingersol, and Ross
(2003). On the one hand, this co-investing may improve the alignment of interests on
the downside but can also lead to overconservatism by the manager on the upside, as
shown in Carpenter (2000). This outcome is often the case because most hedge fund
investors are far more diversified than the hedge fund manager. Of course, the existence of incentive fees mitigates part of this problem. However, precisely how these
two features of the hedge fund compensation model should be combined must ultimately depend on the inherent risk of the strategies used by the fund manager.
Expressed differently, once again the solution to this problem needs the proper identification of the risk factors underlying different hedge fund strategies.

Concluding Remarks
This paper has provided an overview of the growth of the hedge fund industry over the
last decade as it evolved into adolescence. Academic research has played a contributory role in publicizing some of the key features of this opaque industry. However,
research has a way of uncovering more questions than answers. This is clearly the case
with a dynamic, multifaceted industry like hedge funds. By putting forward a simple
model of the hedge fund business, we have pulled together some of the important
issues involving investors in hedge fund products, financial intermediaries, and regulators into a single framework. This framework reveals a fundamental question common
to a number of important concerns regarding the future of the hedge fund industry—
namely, the identification of systemic risk factors inherent in hedge fund strategies.
We believe that this identification is the key input to important questions such as
optimal contract design between buyers and sellers of hedge fund products. These
questions in turn have important implications for risk monitoring of hedge funds by
financial intermediaries as well as regulators.
In addition, understanding these risk factors helps to clarify seemingly complex
and chaotic changes in the hedge fund industry. For example, during the early 1990s,
the hedge fund industry was dominated by macro hedge funds. Today, database

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

vendors report a myriad of hedge fund strategies that have no uniform definition. In
reality, our simple hedge fund business model implies that hedge fund managers will
diversify in order to maximize the enterprise value of their firm. This implication is
consistent with the growing trend of multistrategy hedge funds. There are clear similarities between the way in which macro funds choose factor bets and the tactical
strategy allocation process of multistrategy
The identification of systemic risk factors
hedge funds. Interestingly, both exhibit
significant factor correlation to our risk
inherent in hedge fund strategies is the key
factor model. We would argue that, from
input to important questions such as optimal this perspective, the evolution of hedge
contract design between buyers and sellers
fund strategies has been much more linear
than it may appear at first glance. Apparent
of hedge fund products.
style changes over the last decade are
consistent with hedge fund managers’ maximizing their enterprise value by diversifying the impact of differing life cycles of hedge fund strategies.
The past few years have witnessed acquisition of hedge fund firms by regulated
financial institutions. This trend, coupled with our model of the hedge fund business,
implies that the price discovery process of a hedge fund firm is emerging. From what
we have seen thus far, pricing favors those with a steady, diversified stream of fee
income. This development in turn should reduce the risk of excessive risk taking by
individual hedge fund managers.
However, diversifying behavior by individual hedge funds does not preclude the
risk of leveraged opinions converging onto the same set of bets. Consequently, risk
monitoring of the hedge fund industry should reorient its focus away from megafirms
to the convergence of factor bets. For example, many different strategies may individually appeal to valid reasons to take on credit risk, but, taken together, they may
put the industry as a whole at a dangerous level of concentration on a single risk
factor. The management of convergence risk begins with the identification of risk factors common to different hedge fund strategies.
Recognizing potential risk concentration does not in itself remedy the situation.
More than likely the prevention of convergence risk involves action by hedge fund
investors, financial intermediaries, regulators, and the hedge fund managers themselves. To this end, better transparency will help investors reshape their portfolios
away from excessive exposure to factor bets. Compensation contracts that help identify “me too” types of hedge funds will precipitate withdrawal of capital, thereby
reducing the risk of herding. Fee structures that reward managers on a risk-adjusted
basis will help steer hedge funds away from unwarranted factor bets that could damage their firms’ enterprise value. Identifying risk factors inherent in hedge fund
strategies is a good beginning, but guidance from regulators toward better disclosure
and transparency from hedge fund managers is needed to align everyone’s interest.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

REFERENCES
Ackermann, Carl, Randall McEnally, and David
Ravenscraft. 1999. The performance of hedge funds:
Risk, return, and incentives. Journal of Finance 54,
no. 3:833–74.

Fung, William, and David A. Hsieh. 1997. Empirical
characteristics of dynamic trading strategies: The case
of hedge funds. Review of Financial Studies 10,
no. 2:275–302.

Agarwal, Vikas, William Fung, Yee Cheng Loon, and
Narayan Naik. 2006. Risk and return in convertible
arbitrage: Evidence from the convertible bond market.
European Finance Association 2006 Zurich Meetings
(August 25). Available at <http://papers.ssrn.com/
abstract=889009> (November 19, 2006).

———. 1999. A primer on hedge funds. Journal of
Empirical Finance 6, no. 3:309–31.
———. 2000a. Measuring the market impact of hedge
funds. Journal of Empirical Finance 7, no. 1:1–36.

Agarwal, Vikas, and Narayan Naik. 2004. Risks and
portfolio decisions involving hedge funds. Review of
Financial Studies 17, no. 1:63–98.

———. 2000b. Performance characteristics of hedge
funds and commodity funds: Natural vs. spurious biases.
Journal of Financial and Quantitative Analysis 35,
no. 3:291–307.

Berk, Jonathan B., and Richard Green. 2004. Mutual
fund flows and performance in rational markets.
Journal of Political Economy 112, no. 6:1269–95.

———. 2001. The risk in hedge fund strategies: Theory
and evidence from trend followers. Review of Financial
Studies 14, no. 2:313–41.

Brown, Stephen J., William N. Goetzmann, and Roger G.
Ibbotson. 1999. Offshore hedge funds: Survival and
Performance, 1989–95. Journal of Business 72,
no. 1:91–117.

———. 2002. Risk in fixed-income hedge fund styles.
Journal of Fixed Income 12, no. 2:6–27.

Brown, Stephen J., William N. Goetzmann, Roger G.
Ibbotson, and Stephan A. Ross. 1992. Survivorship
bias in performance studies. Review of Financial
Studies 5, no. 3:553–80.
Brunnermeier, Markus, and Stefan Nagel. 2004. Hedge
funds and the technology bubble. Journal of Finance
59, no. 5:2013–40.
Carpenter, Jennifer. 2000. Does option compensation
increase managerial risk appetite? Journal of Finance
55, no. 5:2311–31.
Chan, Nicholas, Mila Getmansky, Shane Haas, and
Andrew W. Lo. 2006. Systemic risk and hedge funds.
Federal Reserve Bank of Atlanta Economic Review
91, no. 4:49–80.
Duarte, Jefferson, Francis Longstaff, and Fan Yu. 2005.
Risk and return in fixed income arbitrage: Nickels in
front of a steamroller? University of California at Los
Angeles, Anderson Graduate School of Finance
Working Paper 8-04.
Edwards, Franklin. 2006. Hedge funds and investor
protection regulation. Federal Reserve Bank of Atlanta
Economic Review 91, no. 4:35–48.
Eichengreen, Barry, Donald Mathieson, Sunil Sharma,
Bankim Chadha, Laura Kodres, and Anne Jansen.
1998. Hedge Fund and Financial Market Dynamics.
International Monetary Fund Occasional Paper
No. 166, May.
Fama, Eugene, and Kenneth French. 1992. The crosssection of expected stock returns. Journal of Finance
47, no. 2:427–65.

———. 2003. The risks in hedge fund strategies: Alternative alphas and alternative betas. In The new generation of risk management for hedge funds and
private equity funds, edited by Lars Jaeger, 72–87.
London: Euromoney Institutional Investor PLC.
———. 2004a. Extracting portable alphas from equity
long-short hedge funds. Journal of Investment Management 2, no. 4:57–75.
———. 2004b. Hedge fund benchmarks: A risk-based
approach. Financial Analysts Journal 60, no. 5:65–80.
______. 2006. The risk in hedge fund strategies: Theory
and evidence from long/short equity hedge funds. Duke
University working paper.
Fung, William, David A. Hsieh, Narayan Naik, and Tarun
Ramadorai. 2006. Hedge funds: Performance, risk and
capital formation. American Finance Association 2007
Chicago Meetings Paper (July 19). Available at
<http://ssrn.com/abstract=778124> (November 17, 2006).
Fung, William, David A. Hsieh, and Konstantinos
Tsatsaronis. 2000. Do hedge funds disrupt emerging
markets? Brookings-Wharton Papers on Financial
Services, 377–421.
Getmansky, Mila, Andrew W. Lo, and Igor Makarov. 2004.
An econometric model of serial correlation and liquidity
in hedge fund returns. Journal of Financial Economics 74, no. 3:529–609.
Goetzmann, William N., Jonathan E. Ingersol, and
Stephen A. Ross. 2003. High-water marks and hedge
fund management contracts. Journal of Finance 58,
no. 4:1685–1718.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Goldman, M. Barry, Howard Sosin, and Mary Ann Gatto.
1979. Path dependent options: “Buy at the low, sell at
the high.” Journal of Finance 34, no. 5:1111–27.
Krail, Robert, Clifford Asness, and John Liew. 2001. Do
hedge funds hedge? Journal of Portfolio Management
20 (Fall): 6–19.
Lehmann, Bruce. 2006. Corporate governance and
hedge fund management. Federal Reserve Bank of
Atlanta Economic Review 91, no. 4:81–91.
Liang, Bing. 2000. Hedge funds: The living and the
dead. Journal of Financial and Quantitative
Analysis 35, no. 3: 309–26.
Malkiel, Burton. 1995. Returns from investing in equity
mutual funds, 1971 to 1991. Journal of Finance 50,
no. 2:549–72.
Merton, Robert C. 1981. On market timing and investment performance: I. An equilibrium theory of value
for market forecasts. Journal of Business 54,
no. 3:363–407.
Mitchell, Mark, and Todd Pulvino. 2001. Characteristics of risk in risk arbitrage. Journal of Finance 56,
no. 6:2135–75.
Patton, Andrew. 2004. Are “market neutral” hedge funds
really market neutral? EFA 2004 Maastricht Meetings
Paper No. 2691 (March 11). Available at <http://ssrn.
com/abstract=557096> (November 17, 2006).

ECONOMIC REVIEW

Fourth Quarter 2006

Posthuma, Nolke, and Pieter J. van der Sluis. 2003. A
reality check on hedge fund returns. (July 8). Available
at <http://ssrn.com/abstract=438840> (November 17,
2006).
Presidential Task Force on Market Mechanisms. 1988.
Report of the Presidential Task Force on Market
Mechanisms. Washington, D.C., January.
President’s Working Group on Financial Markets. 1999.
Hedge Funds, Leverage, and the Lessons of LongTerm Capital Management: Report of the President’s Working Group on Financial Markets. April.
Schneeweis, Thomas, and Richard Spurgin. 1998.
Multifactor analysis of hedge fund, managed futures,
and mutual fund return and risk characteristics.
Journal of Alternative Investments 1, no. 2:1–24.
Securities and Exchange Commission (SEC). 2003.
Implications of the Growth of Hedge Funds, Staff
Report to the United States Securities and Exchange
Commission, September.
———. 2004. Registration Under the Advisers Act of
Certain Hedge Fund Advisers, 17 CFR Parts 275 and
279, Release No. IA-2333, File No. S7-30-04.
Sharpe, William. 1992. Asset allocation: Management
style and performance measurement. Journal of
Portfolio Management 18, no. 2:7–19.

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Hedge Funds and Investor
Protection Regulation
FRANKLIN R. EDWARDS
The author holds the Arthur F. Burns Chair in Free and Competitive Enterprise and is
director of the Center for the Study of Futures Markets at the Graduate School of Business
at Columbia University. This paper was presented at the Atlanta Fed’s 2006 Financial
Markets Conference, “Hedge Funds: Creators of Risk?” held May 15–18.

n September 29, 2003, the Securities and Exchange Commission (SEC, or the
Commission) issued a report on the Implications of the Growth of Hedge
Funds (SEC 2003). The report raised several concerns related to hedge funds and
proposed a number of regulatory initiatives that the SEC might take. Its principal recommendation was that most hedge managers (advisers) be required to register with
the SEC as investment advisers under the Investment Advisers Act (IAA) of 1940, as
amended. On July 14, 2004, after a lengthy period of public comment on the report,
the SEC adopted (by a three to two vote) Rule 203(b)(3)-2, requiring the registration
of most hedge fund advisers by February 1, 2006.
The Commission’s rationale for adopting Rule 203(b)(3)-2 was that SEC registration of advisers was necessary to protect “investors in hedge funds, and to enhance
the Commission’s ability to protect our nation’s securities markets” (SEC 2004b).
Pursuant to its investor protection mandate, the SEC cited two concerns: the growing
incidence of fraudulent activity by hedge fund advisers and the increasing “retailization”
of hedge funds. Requiring hedge fund advisers to register, the Commission argued,
would deter fraudulent conduct by providing the SEC with better information about
the activities of hedge fund advisers, by giving the SEC the authority to conduct on-site
examinations of hedge fund advisers, and by fostering a standard of conduct and an
environment of compliance that would serve to better protect hedge fund investors.
The Rule 203(b)(3)-2 initiative is not the first time the SEC has expressed concern about the activities of hedge funds. More than thirty years ago, in 1969, the SEC
instituted an investigation of the use of leverage and short selling by hedge funds
(SEC 1969, 18). In 1972 it conducted a study of the use of hedge funds by institutional
investors (SEC 1972, xv). In 1992 the Commission provided Congress with an analysis of the regulatory treatment of hedge funds under the federal securities laws. And
in 1999, in the wake of the September 1998 near collapse of Long-Term Capital
Management Inc. (LTCM) and the intervention of the Federal Reserve in arranging a

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

creditor bailout of LTCM, the Commission, along with other government agencies,
participated in the President’s Working Group on Financial Markets, which examined
the potential impact of hedge funds on the stability of financial markets (President’s
Working Group on Financial Markets 1999).
The concerns about hedge funds manifested by these regulatory initiatives can be
grouped under two general public policy issues: financial stability and investor protection. Specifically, are the activities of hedge funds a threat to financial stability (or do
hedge funds pose a systemic risk)? And, second, are the legal or regulatory protections
for hedge fund investors adequate? The focus of this paper is on the second issue:
investor protection regulation. The discussion first considers the conceptual role of
investor protection regulation: What are its goals and likely costs and benefits? The
following sections describe alternative regulatory approaches to investor protection
and the current regulatory structure under which hedge funds operate in the United
States. A discussion follows of the new regulatory requirements associated with SEC
registration of hedge fund advisers and the purpose of these requirements. The next
sections examine the SEC’s concern about the “retailization” of hedge funds and discuss the role of registered hedge funds and particularly funds of hedge funds (FoFs).
The final sections examine the implications of increased investments in hedge funds
by institutional investors such as pension funds and provide my key conclusions and
suggestions for future regulatory initiatives.

Social Calculus of Investor Protection
In an ideal world all investors would have free access to all investment products and
would have the right to decide for themselves which products to buy or which provided them with the best combinations of risk and return. Investors would make
investment decisions taking into account their current and expected income, their
current portfolio of assets and obligations, and their own tolerance for risk. In response,
asset providers would provide an array of investment products that satisfied the
needs of all investors. In this world investors would be solely responsible for their own
miscalculations and for whatever bad luck they might encounter related to their
investment decisions.
In the real world, of course, things are more complicated. Not all investors have
the same information and are equally capable of knowing how to evaluate whatever
information they do have, and not all vendors of investment products are honest and
straightforward in their dealings with investors. Further, product providers typically
know more about the products they offer than do consumers of that product, and
providers often have an economic incentive not to communicate all of their information to consumers. And even if providers do wish to communicate all of their information
fully, doing this credibly can sometimes be difficult.
These real-world complexities—information asymmetries, potential conflicts of
interest, and disparate investor capabilities—are well understood by many investors,
particularly more financially sophisticated ones, but are clearly not understood by all
retail investors. A common solution for these market complexities is the intermediation of professional investment advisers, whom investors can retain to represent
their interests and to advise them about the most appropriate investments for them.
These advisers can be expected to know more about the investment products being
offered than their investor-clients and should therefore be better able to protect their
clients’ interests. Alternatively, less knowledgeable investors can place their money
with professional fund managers, who can make the appropriate investment decisions for them.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

But these market intermediaries are unlikely to eliminate completely information
asymmetries and disparities of investor sophistication, and they may create additional information problems. Investment advisers, while knowing more than many
investors, are still unlikely to know as much as product providers. They also are likely
to be more motivated to sell their own products than to protect the long-run interests of their clients (even recognizing the potential loss of “reputational capital” of
not doing so). The solution of imposing legal liability on advisers and fund managers
for inappropriate recommendations or investment decisions (such as suitability
requirements) does not fully solve the
problem because of the difficulty of draftIn an ideal world all investors would have
ing and enforcing effective and practicable
the right to decide for themselves which
legal standards.
products to buy or which provided them with
Thus, in the real world governments
and regulators are left to decide how to deal
the best combinations of risk and return.
with these market complexities, or how to
balance the perceived costs of doing nothing to protect investors against the perceived
benefits of proactive intervention to protect them. Government intervention itself is
costly. Direct administrative and bureaucratic costs are associated with regulation,
compliance, market rigidities due to regulatory barriers or prohibitions, distortions of
economic incentives as market participants “game” the regulations, and, possibly, political costs as various private interest groups vie with one another to capture regulators
and shape the regulatory agenda. Also, a potentially large cost of investor protection
regulation is that it may preclude a certain class of investors from participating in
certain investment products, relegating all members of this class to less desirable
investment products.
Hedge funds are an example. Unencumbered by regulatory restrictions on short
selling, leverage, and fee arrangements and by liquidity requirements and portfolio
distributions constraints, hedge funds can use trading strategies not typically available to retail investors, relegating these investors largely to investment products provided by mutual funds. As a result, hedge funds may be able to provide investors with
better downside protection against precipitous falls in asset (stock) prices, such as
occurred in early 2000, than are mutual funds, which typically hold long equity or
bond positions and cash. Thus, blocking retail investor access to hedge funds may
impose significant costs on the excluded investors by forcing them into inferior
investment products, which must be balanced against the potential benefits of protecting investors against losses they might incur if they were to invest in hedge funds.1
The difficulty of quantifying these competing considerations has resulted in different
countries reaching different conclusions about how best to balance these interests.
But in almost all countries the issue has been resolved in favor of regulation to protect some, if not all, retail investors, although not always in the same way. Implicitly,
most countries have judged that the social costs associated with real-world market
complexities (or imperfections) are greater than the costs of regulatory intervention,
or, put another way, that the potential benefits of regulation are greater than the
potential costs. Not everyone can be expected to agree with this social calculus, and
some will argue that all government decisions are driven only by the political power
of special-interest groups rather than considerations of social or economic welfare.
But for purposes of this paper I accept that some investor protection regulation will
1. For a discussion of the investment performance of hedge funds, see Edwards and Gaon (2003); for
a comprehensive survey of past research on hedge fund performance, see Naik and Aggarwal (2005).

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

always be an integral component of retail investment markets in most countries and
that the more relevant question is how best to shape investor protection regulation.

Alternative Investor Protection Regimes
Investor protection regulatory regimes in most countries can be described as either
“top down” or “bottom up.” A top-down regime is characterized by the requirement
that investment products or schemes be authorized together with rules about what
that scheme can and cannot do. For example, the regulation of investment companies
(mutual funds) in the United States is primarily a top-down structure. Mutual funds
must register under the Investment Company Act (ICA) of 1940 and must adhere
to detailed SEC regulations with respect to custodial requirements, liquidity and
diversification portfolio requirements, restrictions on leveraging and short selling,
management fee arrangements, redemption requirements, disclosure and reporting requirements, and so forth. The primary purpose of these regulations is both to
better inform investors and to protect them by limiting exposure to financial loss.
The regulatory regimes for listing and authorization of investment funds in the
United Kingdom are predominately top-down, as is much European legislation governing investment funds.2
In contrast, a bottom-up regulatory regime is basically a disclosure-based regime.
Greater reliance is placed on rules that require investment product providers to
accurately describe the nature of these investment products and their potential risks.
Armed with this information, investors are given much more responsibility to assess
the risks and to determine whether the investments are suitable for them. Fundamental
to this scheme is an acceptance on the part of both regulators and investors that
some investment products will fail and that some investors will experience significant
financial losses, perhaps even their entire investments. Australia’s regulatory scheme
for mutual funds is an example of this approach. As a result, in Australia a very wide
range of retail funds exists, including hedge funds and other exotic funds, such as
raptor funds that invest in ostrich farms.

U.S. Regulation of Hedge Funds
In the United States the regulation of hedge funds might be best characterized as a
patchwork of exemptions from various investor protection laws rather than a
thoughtfully crafted top-down or bottom-up regulatory scheme. The United States neither requires government authorization of hedge funds nor restricts what hedge funds
are able to do, nor does it mandate that hedge funds and hedge fund advisers (prior to
this year) make specific disclosures to investors. But to gain these exemptions hedge
funds must restrict their clients to investors who meet certain threshold wealth or
income requirements. Qualifying investors are given unlimited access to hedge funds
with virtually no regulatory protections, while low-wealth or low-income investors are
entirely excluded from participating in any kind of hedge fund investment.
Thus, in the United States we have, perhaps unwittingly, separated hedge fund
investors into two distinct classes: retail (investors who do not meet the threshold
wealth requirements for exemption) and wholesale (investors who do meet the
threshold wealth requirements). Exactly where the legal threshold levels of wealth
and income are set determines which investors are retail and which are wholesale.
These legal thresholds, it should be noted, are not specific to hedge funds and were
established years ago, when hedge funds were not part of the financial landscape.
Rather, their purpose was to determine the disclosure obligations applicable to issuers
in public versus private securities and the scope of mutual fund regulation.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

To be more precise, hedge funds are investment pools and are potentially subject
to a variety of legal restrictions and regulations unless they are organized in a way
that exempts them from these regulations, specifically from the 1933 and 1934
Securities Acts, the ICA, and the IAA. Hedge funds are exempt from the Securities Act
of 1933 if they obtain their investors through private placements rather than a public
offering. This exemption hinges on meeting the requirements of section 4(2) or
Regulation D of the 1933 act; it usually means restricting the fund’s investors to
“accredited” investors. If hedge funds fail to meet this test, they would have to file a
registration statement under the 1933 act, which would require that extensive information be disclosed and would create liability for material misstatements or omissions.3
Accredited investors are individuals who have incomes of at least $200,000 in
each of the two most recent years, who have a joint income with a spouse in excess
of $300,000 in each of those years (and who have a reasonable expectation of reaching the same income level in the current year), or who have a net worth, or joint net
worth with a spouse, that exceeds $1 million at the time of purchase. Institutional
investors with assets in excess of $5 million, banks, savings and loan associations,
broker/dealers, insurance companies, investment companies, and small business
investment companies licensed by the U.S. Small Business Administration are also
accredited investors. The rationale for limiting investor access to private placements
to accredited investors is that such investors can be assumed to be both informed
and sophisticated enough not to need the protections afforded to other investors
under the federal securities laws.
Hedge funds typically exempt out of the 1934 Securities Act by limiting their
investors to fewer than 500. If a fund has more than $10 million and 500 investors, its
securities would have to be registered with the SEC under the 1934 act, and it would
become a “reporting” company. This classification would mean providing investors
with extensive disclosure: annual reports (Form 10-K), quarterly reports (Form 10-Q),
and so on.
Most hedge funds also exempt out of the ICA, which regulates mutual funds, by
relying on the exceptions in either section 3(c)(1) or 3(c)(7) of the act. Under 3(c)(1),
the act does not apply to an investment pool (or hedge fund) that does not obtain its
investors through a public offering (and is therefore exempt from the 1933 Securities
Act) and has fewer than 100 persons or legal entities (investors). Under section
3(c)(7), a hedge fund is exempt from the act if it has only investors who meet the
criterion of qualified purchasers—individuals or companies who have at least $5 million in investments. The qualified-purchaser threshold is considerably higher than
the accredited-investor threshold.4
If a hedge fund were not to exempt out of the ICA, it would have to file as a “registered investment company” under the act and would become subject to numerous
top-down regulations governing its portfolio holdings, leverage, short selling, marketing, governance, and conflict-of-interest and disclosure rules. Because of the
nature of most hedge funds’ investment strategies, such registration would most likely
take the form of a closed-end fund, relieving the fund of the same reporting and
redemption requirements imposed on open-end mutual funds.
2. For an overview, see the Financial Services and Markets Act 2000 and a council directive of the
European Commission (1985), as amended.
3. While Rule 506 of the 1933 act does allow them to have as many as thirty-five “nonaccredited”
investors, most hedge funds find it is not worthwhile to involve themselves with such investors.
4. See the ICA, sec. 2(1)(51), and SEC Rule 2a 51-1.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

As exempt investment pools, hedge funds can trade any type of security or financial instrument, operate in any market anywhere in the world, make unlimited use of
derivatives instruments, engage in unrestricted short selling, employ unlimited amounts
of leverage, hold concentrated positions in any security without restriction, set their
own redemption policies without restriction, and use whatever fee structure seems
most productive to compensate their managers or advisers. Nor are exempt pools
required to make extensive disclosures to investors or regulators. Thus, limits or
restrictions on hedge funds’ activities are determined not by regulation but primarily
by the contractual relationships they have with their investors and by market discipline
exerted by the creditors, counterparties,
and investors with whom they transact.
In the United States we have, perhaps
In February 2006 the SEC adopted
unwittingly, separated hedge fund
Rule 203(b)(3)-2, which introduced a topinvestors into two distinct classes:
down approach to regulating the activities
of hedge funds by requiring SEC registraretail and wholesale.
tion of most hedge fund advisers. The rule’s
intent is to provide greater protection for hedge fund investors through enhanced
disclosure and increased regulatory oversight of the activities of hedge fund advisers.5
Until this year advisers to hedge funds did not have to register under the IAA
because they typically met the private-adviser exemption. Specifically, if they had
fewer than fifteen clients in the past twelve months and did not hold themselves out
to the public as an investment adviser or act as an investment adviser to a registered
investment company or business development company, they were not required to
register (Rule 203[b][3]). Under previous SEC rules each separate company (hedge
fund, investment partnership, managed account, etc.) that the adviser managed was
considered a single client for purposes of registration if the adviser based its advice
to the company on the company’s investment objectives as opposed to the investment
objectives of the company’s individual owners.
Rule 203(b)(3)-2 changes this definition of “client” and by doing so effectively
requires all hedge fund advisers to register with the SEC (except those with no more
than $25 million under management or whose funds require a “lockup” longer than
two years.)6 This rule change was accomplished by introducing a “look-through”
provision that requires each owner (investor) of a private fund to be counted as a
client. Under the new rule, an adviser to a single hedge fund with fifteen or more
clients (investors) must register under the IAA. Further, Rule 203(b)(3)-2(k) requires
advisers to hedge funds in which registered investment companies (mutual funds)
invest to count all investors in those mutual funds as clients and to look through
the top fund in a fund-of-fund structure and count as clients all investors in the
portfolio hedge funds.
To limit the reach of Rule 203(b)(3)-2 to advisers of hedge funds, however, the
Commission defines a hedge fund for purposes of registration to be a company or fund
that (1) would be subject to regulation under the ICA but for the exceptions provided
by either section 3(c)(1) or section 3(c)(7) of the act; (2) permits its investors to
redeem their interests in the fund within two years of purchasing them (that is, have
less than a two-year lockup); and (3) is offered with investment strategies that are
based on the skills, ability, or expertise of the investment adviser. This definition is
intended to exclude advisers to many other business organizations, such as insurance
companies, broker-dealers, and banks, as well as advisers to private equity funds and
venture capital funds that typically require lockup periods longer than two years.
While the SEC acknowledges that private funds such as private equity and venture

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

capital funds are very similar to hedge funds, it defends its definition on the grounds
that it has not encountered significant enforcement problems with advisers to such
funds, in contrast to its experience with hedge fund advisers.
The import of requiring hedge fund advisers to register under the IAA is that they
become subject to much the same rules that apply to advisers to mutual funds (registered investment companies). Specifically, hedge fund advisers are subject to examination by the SEC, conflict-of-interest and antifraud rules, additional disclosure
requirements, and limitations on the use of performance fees (Section 205[a][1]).
However, if all of the advisers’ clients are qualified clients (have a net worth of more
than $1.5 million or $750,000 invested with the adviser), registered advisers can employ
an asymmetric fee structure.7 Because almost all hedge fund advisers will want to
use asymmetrical performance-based fee structures, an indirect effect of Rule
203(b)(3)-2 is to raise the minimum net worth requirement for 3(c)(1) hedge funds
to that of the qualified client standard from the prior accredited investor standard
(net worth of $1 million or annual income of more than $200,000).8

What Will Registration of Hedge Fund Advisers Accomplish?
While a comprehensive description of the many regulatory requirements that accompany adviser registration is beyond the scope of this paper, a description of a few key
requirements may provide a flavor of the nature of these regulations. Registered
hedge fund advisers must complete the Uniform Application for Investment Adviser
Registration (or Form ADV) under Rule 204-3. Part I of the ADV requires information
about the adviser’s business location, ownership structure, basic operations, and past
disciplinary events. Part II requires information about the adviser’s fees, investment
style, potential conflicts of interest, brokerage practices, affiliations with other securities
professionals, education and business background, and other information relevant to
a client’s decision to hire the adviser. Part I is made available on the Internet. Part II
must be provided to clients and is, in effect, a mandated disclosure document. Unless
5. In the absence of SEC registration, hedge fund advisers are still subject to antifraud provisions of
both federal securities laws and state antifraud rules, such as New York’s Martin Act, as well as to
insider trading laws. Eliot Spitzer used the Martin Act to successfully prosecute widespread mutual
fund fraud during the last few years, long before the SEC acted against these abuses.
6. This lockup exemption is intended to exempt advisers to private equity funds and venture capital
funds from registration.
7. In the absence of this exception, performance-based fee structures for mutual fund advisers must
be fulcrum fees that move in both directions equally, as opposed to the asymmetric fees that are
common in hedge funds (for example, 20 percent of net returns above some “hurdle rate”).
8. Many hedge funds also are regulated by the Commodities Future Trading Commission as “commodity pool operators” (CPOs) because they invest in or trade one or more futures or options
contracts on a regulated commodity exchange. The Commodity Exchange Act (CEA) subjects CPOs
and their advisers (CTAs) to regulation but not the commodity pools themselves. Once registered,
CPOs and CTAs must comply with the rules of the National Futures Association (NFA), avoid conflicts
of interest and protect customer funds, provide written disclosure to prospective investors of
the risks of investing in commodity interests, adhere to restrictions on advertising, satisfy recordkeeping and reporting requirements, and be subject to periodic inspections of their activities by
the NFA. In addition, advisers to hedge funds are subject to common law remedies for fraud, as
well as claims for fraudulent manipulation under section 10(b) and Rule 10b-5 of the Securities
Act of 1934. Typically, prior to investing in a hedge fund in a private placement, investors are given
for review and agreement an offering memorandum and partnership agreement. These documents
provide investors with information about the potential risks associated with the fund and serve as a
notice of caveat emptor. They also form the basis for possible contractual law and fraud remedies
available to investors.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

the adviser utilizes a qualified custodian, it must send quarterly account statements
to clients, deliver an audited balance sheet with Form ADV Part II, and undergo an
annual surprise audit by an independent certified public accountant (CPA).
Additional disclosure is required for registered advisers if they experience an
impaired financial situation (if advisers’ financial condition is “reasonably likely to impair
the ability of the adviser to meet contractual commitments to clients” or if any legal or
disciplinary events occur that are material to an evaluation of the advisers’ integrity or
ability to meet contractual commitments to clients (Rule 206[4]-4). Further, Section 207
of the IAA makes it unlawful for a registered investment adviser to willfully make an
untrue statement of a material fact, or to willfully omit to state such a material fact, in
any registration application or report filed with the SEC.
Registration carries significant record-keeping requirements for advisers. For
example, advisers must keep copies of all disclosure documents given to clients or
prospective clients and the dates that each disclosure was given or offered; retain
performance data and supporting calculations and work papers related to the performance data used; retain client transaction records (securities purchased, sold, the
date, amount, price) and client securities position listings; retain client suitability
documentation, such as basic information on the client; and maintain performance
records for the five-year period from the date last used. Additional record-keeping
requirements apply to insider trading policies and reportable securities transactions
by so-called access persons.
Finally, if hedge fund advisers meet the test of “having custody of client assets,”
as many will do, they will have to maintain client funds and securities with a qualified
custodian, such as a bank or registered broker-dealer. The qualified custodian is
required to deliver account statements (on at least a quarterly basis) directly to the
client to ensure the integrity of the statement and enable the client to identify any
erroneous or unauthorized transactions or withdrawals by the adviser. If an adviser
does retain a qualified custodian, it does not have to send quarterly statements to
clients or undergo an annual surprise examination by an independent CPA to verify
the funds and securities of the clients.9
The obvious thrust of Rule 203(3)-2 is to provide greater protection for hedge
fund investors against fraudulent activities by hedge fund advisers. Little in the new
disclosure requirements will assist hedge fund investors in evaluating the nature of
the hedge fund investments or the risks associated with those investment strategies.
Thus, Rule 203(3)-2 will not assist investors in evaluating the likely performance of
a particular hedge fund investment strategy or in comparing that investment with
alternative investment products.
Critics of Rule 203(3)-2 argue that registration of hedge fund advisers is unlikely
to provide effective protection against fraud. But even assuming that there will be
some reduction in fraud losses to hedge fund clients, critics contend that the costs
(direct and indirect) associated with registration are likely to be considerably greater
than the benefits of greater protection for hedge fund investors, who are considerably wealthier and more financially literate than the average mutual fund investor.
The costs associated with increased regulation include added SEC costs and compliance costs to hedge funds that will be passed on to investors, reducing their returns.
Critics also find little reason to believe that there are significant negative externalities
(or social costs) associated with private investment losses incurred by wholesale
hedge fund investors.10 Such losses are unlikely to undermine confidence in financial
markets generally or result in contagion effects that undermine other financial institutions. Finally, Rule 203(3)-2 will shut out more investors from hedge fund invest-

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

ments by effectively raising the minimum wealth threshold for hedge fund investors
to that of the qualified-client standard, arguably relegating them to inferior investment vehicles.

The Retailization of Hedge Funds
In supporting Rule 203(3)-2 the SEC expressed concern about the growth of what it
termed the retailization of hedge funds—the increasing ability of less qualified (or
retail) investors to access hedge fund investments. It pointed to three ways that this
increased access was happening (SEC 2004a).
First, the wealth thresholds that restrict investor access to hedge funds, such as
the accredited investor standard applicable to 3(c)(1) hedge funds, have been eroded
over time by a general rise in income and wealth levels. For example, a recent report
indicated that the number of American households with a net worth of $1 million or
more, excluding their principal residence, grew from 5.2 million in 2002 to 8.9 million
in 2005 (Johnston 2006). If principal residence were not excluded this number would
be two or three times greater. Thus, a far larger segment of the investing public is probably now able to meet the $1 million accredited-investor standard necessary to access
hedge funds than when this standard was established. The issue, however, is whether
someone with $1 million today is less financially sophisticated than in the past, which
is not obvious. But it is possible that more unsophisticated investors are able to participate in hedge fund investments than in prior years and that this participation may have
contributed to the growing fraud problem that the SEC has observed. (The underlying
assumption, of course, is that a reasonably close positive correlation exists between an
individual’s financial wealth and his or her financial sophistication.)11
Whatever the truth about this relationship, the adoption of Rule 203(3)-(2) indirectly redresses this concern by in effect raising the minimum wealth threshold for
3(c)(1) hedge fund investors to that of the qualified-client standard. Specifically,
because most hedge fund managers will want to use an asymmetric performancebased fee structure, the minimum wealth standard for individual investors for 3(c)(1)
funds will now be a net worth of at least $1.5 million rather than the $200,000 annual
income or $1 million net worth thresholds under the accredited-investor standard.12
Second, there has been a proliferation of funds of hedge funds (FoFs), which has
arguably made hedge fund investments more available to more retail investors
because FoFs typically have lower investment minimums for individuals. FoFs are
hedge funds that invest only in other hedge funds—the portfolio funds—or hold participations (or are limited partners) in the portfolio hedge funds. The appeal of FoFs
to investors is diversification and professional management. FoFs provide diversification benefits by investing in many different hedge funds, thereby diversifying across
9. Registered hedge fund advisers, unlike mutual fund advisers, do not have to file quarterly reports
with the SEC listing all the securities they own, file semiannual reports to shareholders about
their operations, disclose how they vote their proxies, have a capital structure that allows only
one class of stock to be issued, or have a board with an independent chairman and a majority of
independent directors.
10. SEC registration may result in a higher incidence of fraud if hedge fund investors become more
reckless in their choice of advisers because they believe SEC oversight now protects them against
such fraud.
11. Recent episodes of fraud suggest that this premise may not be sound. See “NFL players sue a
hedge fund for fraud, theft” (2006).
12. 3(c)(7) hedge funds are not affected because the applicable net worth threshold is already above
the qualified-client threshold.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

risks associated with different hedge fund investment strategies as well as protecting
against the possible fraudulent behavior of hedge fund advisers. In addition, FoF
managers arguably have greater expertise in identifying superior fund managers and
in monitoring their performance than do individual investors, and they may be better
positioned to detect and/or prevent fraudulent behavior on the part of advisers. As a
result, one would expect FoFs to outperform (at least before fees) a passively managed index of diversified hedge fund strategies randomly selected.
These potential benefits do not come free. Investors in FoFs pay another layer of
fees to advisers of FoFs similar to what is typically paid to the advisers of the portfolio
funds. Specifically, FoF investors usually pay FoF managers an annual flat fee of 1 to
2 percent of assets and an incentive fee of, say, 20 percent of net returns above a
threshold level. These fees are in addition to the fees (which are similarly structured)
that the FoF pays to the advisers of each of its portfolio funds. The resulting total
fees are substantial and typically require FoFs to earn before-fee net annual returns
in excess of 40 percent before FoF investors can realize positive net returns. Despite
these fees, FoFs have grown rapidly in recent years.
Third, institutional investors have increased their participation in hedge funds.
While endowments and universities have long been active participants in hedge funds,
more recently pension funds have been increasing their investments in hedge
funds. This trend can be viewed as the indirect retailization of hedge funds because
more pension fund beneficiaries (or retail investors) are indirectly exposed to the
risks associated with hedge fund investments, possibly without any knowledge or
understanding of these risks. Underlying this concern is the implication that the
interests of pension fund managers and advisers may not always be aligned with the
interests of their beneficiaries, which may result in fund managers undertaking
investments that are inappropriate for fund beneficiaries (the principal-agent problem). Further, there is a veiled presumption that the governance structure of pension
funds and other institutional investors cannot be relied on to represent the interests
of fund beneficiaries.

Funds of Funds and Registered Hedge Funds
There are two types of FoFs: unregistered and registered. Unregistered FoFs are
similar to other hedge funds. They are subject to the exemptions from regulation discussed earlier and to the wealth thresholds that apply to investors in hedge funds
generally—the 3(c)(7) qualified-purchaser standard and the qualified-client standard
in effect since the adoption of Rule 203(b)(3)-2. Thus, unregistered FoFs generally
do not have retail investors and as such have not contributed to concern about the
retailization of hedge funds.
Registered FoFs may have retail investors.13 These funds are registered under the
ICA, mostly as closed-end (mutual) funds but sometimes as open-end mutual funds,
and typically pursue an absolute-return investment strategy (such as long/short equity
or market neutral) similar to what unregistered FoFs might utilize. Open-end mutual
funds must honor all redemption requests immediately (or at least within seven days).
Closed-end funds, in contrast, do not issue redeemable securities and may or may not
have publicly traded shares; they may provide their shareholders with liquidity by
agreeing to purchase periodically their own shares at net asset value (so-called interval
funds). At year-end 2002 there were forty-two registered hedge funds, only thirteen
of which had registered their securities under the 1933 Securities Act.
An example is Oppenheimer Tremont’s Market Neutral Fund, which is a registered FoF. The fund pursues a typical FoF investment strategy, offers its securities

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

publicly, and requires a minimum investment of only $25,000. Further, though the
provision is not necessary under the ICA, Oppenheimer Tremont requires its investors
to have a net worth of at least $1.5 million so that its advisers can utilize the standard
hedge fund (asymmetric) incentive fee structure (Rule 205-3).
Registered hedge funds are not a significant threat to the current regulatory scheme
for protecting retail investors. Neither closed-end nor open-end registered mutual
funds are an attractive vehicle through which to pursue most hedge fund investment
strategies. First, current regulations require mutual funds to hold substantial amounts
of liquid assets against possible redemption requests, even in the case of closed-end
interval funds. Because many hedge fund strategies entail holding substantial amounts
of illiquid assets, such a liquidity requirement makes it impossible to profitably purAn analysis of the likely costs and benefits
sue these illiquid strategies.
of increased regulatory protection for hedge
Second, most hedge fund strategies
fund investors suggests that the costs are
rely heavily on the use of leverage, which
is subject to cumbersome restrictions when
likely to be greater than the benefits.
operating as a mutual fund. While the use
of leverage by closed-end funds is less restricted than for open-end funds, both types
of funds are limited in the amount of leverage they may use. Hedge funds are unrestricted in their use of leverage.
Third, most hedge fund strategies employ short selling, which is effectively limited
by mutual fund regulation. Although short selling is not prohibited per se under mutual
fund regulation, the requirement that mutual funds segregate cash and other liquid
securities to cover short positions effectively makes these assets nonproductive and
discourages short selling. Hedge funds have no deterrents to short selling.
Fourth, mutual fund regulation restricts the use of the standard performance-fee
structure used by hedge funds to align the interests of fund advisers and investors.
In particular, hedge funds typically compensate advisers based on the performance
of the fund’s portfolio (typically 20 percent of returns above a designated hurdle rate
or absolute return such as the Treasury bill rate) in order to provide an incentive for
them to produce positive returns in all kinds of market environments (even declining stock markets). In addition, hedge funds usually require advisers to invest their
own assets alongside their investors’ to ensure that they do not engage in excessively
high-risk strategies.
Thus, registration under the ICA is unlikely to attract a great number of hedge
funds because of regulatory restrictions that make it difficult for them to pursue most
hedge fund investment strategies. More generally, Congress and the SEC need to
reexamine these regulatory restrictions to determine whether they are still necessary. Relaxing these restrictions would have two potential benefits: Hedge funds
would have a greater incentive to register under the ICA, and traditional mutual
funds would be able to provide absolute-return investment strategies in competition
with hedge funds. This change could provide retail investors with access to hedge
fund strategies under an acceptable regulatory structure.
The current regulatory structure is, in any case, probably unworkable. While registered FoFs and other hedge funds are subject to the ICA and IAA and to corresponding SEC regulations, like mutual funds, it is not clear how current mutual
regulations can be effectively applied to registered hedge funds. A few examples may
13. Investors in registered funds are not subject to the net-worth or income thresholds applicable to
investors in unregistered hedge funds.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

illustrate this point. First, although mutual funds are subject to substantial disclosure
requirements intended to make their investments and risk exposures transparent to
investors, the same disclosure requirements applied to hedge funds may not provide
much transparency about investment risks and returns. Registered FoFs will typically
not be able to disclose much about the investment strategies pursued by their portfolio hedge funds other than what these funds do generally and the magnitude of the
FoFs’ own investments in each of the portfolio funds. Very little about the nature of
the risks associated with the underlying investment strategies of these funds is likely
to be revealed for proprietary reasons.
Second, the valuations of the FoFs’ portfolio assets and liabilities (or investments)
also are likely to be more challenging than for most mutual funds’ assets, which are
usually traded in liquid public markets. Hedge fund portfolios typically include illiquid assets and complex derivatives trades that make asset valuation considerably more
difficult and inexact. The valuation process may also contain an inherent conflict of
interest because it is often done by hedge fund advisers themselves.
Third, while current mutual fund regulation effectively limits the use of leverage
and short selling by mutual funds, it is not clear how these restrictions would be
effective in limiting the use of leverage and short selling by the portfolio funds of registered FoFs. Unregistered portfolio hedge funds held by registered FoFs would presumably not be subject to these restrictions.
Fourth, the standard reporting requirements directed at mutual fund performance may be inappropriate for registered FoFs as well as for other hedge funds and
may even mislead investors about the true nature of the returns and risks associated
with investments in these funds. For example, conventional performance measures
such as Sharpe ratios can be highly misleading when applied to hedge fund strategies
because of their use of conventional measures of return volatility (such as variance)
as a measure of risk. Such measures do not adequately capture the so-called fat-tail
risks implicit in many hedge fund strategies. Moreover, at present no consensus
exists about how to measure and report hedge fund performance so that investors
will be provided with a clear idea of the likely risks and returns associated with different hedge fund investment strategies (Financial Economists Roundtable 2005).
Thus, there needs to be an open discussion among regulators, industry representatives, academics, and other policymakers about whether all current mutual fund regulations are necessary and whether the same regulations can be effectively applied both to
mutual funds and to registered hedge funds. Potential benefits exist in allowing retail
investors greater access to hedge funds through registered FoFs. Hedge fund investment strategies provide greater diversification opportunities and may result in higher
risk-adjusted returns for investors.14 For example, during the 2000–02 period, when
stock markets were in decline, hedge funds in general performed significantly better
than did stock and bond mutual funds.15 Had retail investors been able to include in their
portfolios some hedge fund investments, they may have been able to avoid or mitigate
the substantial losses that most mutual funds investors incurred during this period.16
The Financial Services Authority (FSA) in the United Kingdom has already
begun this process. It announced recently that it is considering allowing FoFs to
be marketed to retail investors under the same rules that govern mutual funds.
Clive Briault, FSA managing director for retail markets, said, “Given the reality of
the contemporary retail market, it seems sensible to permit the marketing of funds
of hedge funds through authorized, onshore vehicles. These onshore funds of
funds would benefit from the protections already in place for authorized funds”
(“Hedge funds for the masses” 2006).

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Institutional Investors
Another concern has been the growth of investments in hedge funds by institutional
investments, particularly by both private and public pension funds. This concern centers on whether pension fund beneficiaries understand the potential risks associated
with hedge fund investments, and it indirectly raises the principal-agent problems that
may occur when retail investors delegate decision-making authority to professional
fund advisers or managers. In particular, pension fund managers may have conflicts of
interest that are not well understood by the fund’s beneficiaries and that result in the
misalignment of their interests with those of the fund’s beneficiaries. For example,
fund managers may be willing to take more risk than fund beneficiaries would wish in
the hope of increasing the fund’s returns in order to increase their own compensation.
This conflict is not new and is well understood by many investors. Standard solutions have been to try to structure fund-management fees in a way that aligns the
interests of fund managers and beneficiaries or to establish effective institutional
oversight and governance mechanisms to monitor fund managers. In practice, however,
these solutions seldom completely eliminate all conflicts or align the interests of fund
managers and beneficiaries.
Hedge funds are not the source of this conflict, only its latest manifestation.
Hedge funds are attractive to professional fund managers because they hold the
promise of higher risk-adjusted returns, which is especially enticing currently because
traditional asset classes like stocks and bonds are not yielding high returns. As a consequence, some observers have suggested placing restrictions on the ability of pension
funds to invest in hedge funds, such as limiting the hedge fund investments of pension funds to no more than 5 percent of the fund’s assets. Blanket restrictions on hedge
investments by institutional investors, however, would be counterproductive. Any
specific limits would obviously be arbitrary and would suffer from the one-size-fits-all
shortcoming. The problem is not hedge funds but the principal-agent conflicts implicit
in delegating fund management responsibilities to professional fund managers. The
preferable way to address this problem, therefore, is to focus on solutions to that
problem, in particular how to enhance the governance of institutional investors so
that there is more effective oversight of fund managers by fund beneficiaries.

Conclusions and New Regulatory Directions
While a number of concerns have been voiced about the growth of investor interest in
hedge funds, this paper concludes that there is not a strong case for increased regulatory protection for hedge fund investors. An analysis of the likely costs and benefits of
such regulation suggests that the costs are likely to be greater than the benefits. Indeed,
I believe that, on a cost-benefit basis, it is difficult to make a case even for the recently
adopted Rule 203(b)(3)-2. Nonetheless, the most important conclusion of this paper is
14. See Edwards and Gaon (2003, 8–21, table 4), who show that the correlations between stock
returns and the returns on some hedge fund strategies are often low, and Edwards and Caglayan
(2001, 97–108).
15. See Edwards and Gaon (2003, 13, table 3). If the performance of hedge funds pursuing a particular
investment strategy is measured by the average risk-adjusted returns earned by these funds (that is,
their Sharpe ratios), most hedge funds would have outperformed an investment in either the
Standard & Poors 500 stock index or the JP Morgan U.S. Bond Index during the 1990–2002 period.
16. Such restrictions also may distort the flow of investor capital, resulting in market inefficiencies and
reduced liquidity in some markets. Indeed, one explanation for the success of hedge funds is that
they have been able to exploit market inefficiencies in markets that are not mainstream markets.
Restricting the flow of capital into hedge funds therefore may perpetuate these inefficiencies.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

that hedge fund investment strategies should be made more, not less, accessible to a
broader spectrum of investors than at present. In particular, the SEC should consider
authorizing funds of hedge funds under a regulatory structure that better enables hedge
funds to pursue absolute-return strategies so that retail investors can benefit from them.
Part of this process would be to reexamine current mutual fund regulations. The goal of
this process should be to establish a regulatory approach that enables both mutual funds
and hedge funds to provide absolute-return investment strategies to a broader segment
of investors under a regulatory structure that adequately protects investors.
A related proposal was made in a 2003 SEC staff report, which the Commission
refers to in support of its adoption of Rule 203(b)(3)-2. In that report the SEC staff
says, “We recommend that the Commission consider issuing a Concept Release exploring the wider use of hedge fund–type/absolute return strategies. . . . These investments typically have lower correlations to the broader debt and equity markets and
thus may provide benefits to investors under a wider variety of market conditions.
The staff believes that these investments may have benefits that could assist other
investors, including retail investors, in diversifying their overall portfolios. The staff
is not recommending that hedge funds be made more readily available and does not
believe that direct investment into hedge funds by retail investors is appropriate.
Instead we believe it may be the case that retail investors interested in absolute
return strategies should be able to pursue those investments through the registered
investment company structure” (SEC 2003, 103–04).
To my knowledge, the Commission has not followed up on that recommendation.
In my view, the SEC should begin the process of evaluating the need for many of the regulations now imposed on mutual funds by the ICA (such as those pertaining to leverage,
short selling, and liquidity) and should consider developing disclosure requirements
more responsive to the absolute-return strategies now used by many investment funds.

REFERENCES
Edwards, Franklin, and Mustafa Onur Caglayan. 2001.
Hedge fund and commodity fund investments in bull
and bear markets. Journal of Portfolio Management
27, no. 4:97–108.

President’s Working Group on Financial Markets. 1999.
Hedge Funds, Leverage, and the Lessons of LongTerm Capital Management: Report of the President’s
Working Group on Financial Markets. April.

Edwards, Franklin R., and Stav Gaon. 2003. Hedge funds:
What do we know? Journal of Applied Corporate
Finance 15, no. 4:58–71.

Securities and Exchange Commission (SEC). 1969.
35th Annual Report of the Securities and Exchange
Commission.

European Commission. 1985. Council Directive
85/611/EEC of 20 December 1985, O.J.L365,
31.12.85, as amended.

———. 1972. Institutional Investor Study, Report,
H.R. Doc. No. 92-64, 92d Cong., 2d sess.

Financial Economists Roundtable. 2005. Statement on
Hedge Funds, November.
Hedge funds for the masses. 2006. Wall Street Journal,
March 24.
Johnston, David Cay. 2006. New rise in number of millionaire families. New York Times, March 28.
Naik, Narayan, and Vikas Aggarwal. 2005. Hedge funds.
In Foundations and Trends in Finance 1, no. 2.
NFL players sue a hedge fund for fraud, theft. 2006.
Wall Street Journal, February 18–19.

ECONOMIC REVIEW

Fourth Quarter 2006

———. 2003. Implications of the Growth of Hedge
Funds, Staff Report to the United States Securities
and Exchange Commission, September.
———. 2004a. Registration Under the Advisers Act
of Certain Hedge Fund Advisers: Proposed Rule,
17 CFR Parts 275 and 279. Federal Register 69,
no. 144, July 28.
———. 2004b. Registration Under the Advisers Act
of Certain Hedge Fund Advisers, 17 CFR Parts 275
and 279, Release No. IA-2333, File No. S7-30-04.

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Do Hedge Funds Increase
Systemic Risk?
NICHOLAS CHAN, MILA GETMANSKY, SHANE M. HAAS, AND ANDREW W. LO
Chan and Haas are senior research scientists at AlphaSimplex Group, LLC, in Cambridge,
Massachusetts. Getmansky is an assistant professor at the Isenberg School of Management
at the University of Massachusetts. Lo is the Harris & Harris Group Professor of Finance at
the Sloan School of Management at the Massachusetts Institute of Technology, the director
of MIT’s Laboratory for Financial Engineering, and the founder and chief scientific officer
of AlphaSimplex Group, LLC. The authors thank Mark Carey, Kevin Warsh, David Modest,
René Stulz, and participants of the NBER conference on “The Risks of Financial Institutions”
and the Atlanta Fed’s 2006 Financial Markets Conference for helpful comments and discussion and AlphaSimplex Group and the MIT Laboratory for Financial Engineering for
research support. Parts of this article include ideas and exposition from several previously
published papers and books of some of the authors—Getmansky, Lo, and Makarov (2004),
Getmansky, Lo, and Mei (2004), and Lo (2001, 2002). This article is an abridged version of
the paper “Systemic Risk and Hedge Funds,” presented at the Atlanta Fed’s 2006 Financial
Markets Conference, “Hedge Funds: Creators of Risk?” held May 15–18; the longer version
will appear in The Risks of Financial Institutions and the Financial Sector from the
University of Chicago Press in January 2007.

he term “systemic risk” is commonly used to describe the possibility of a series
of correlated defaults among financial institutions—typically banks—that occurs
over a short period of time, often caused by a single major event. A classic example
is a banking panic in which large groups of depositors decide to withdraw their funds
simultaneously, creating a run on bank assets that can ultimately lead to multiple
bank failures. Banking panics were not uncommon in the United States during the
nineteenth and early twentieth centuries, culminating with an average of 2,000 bank
failures per year during the 1930–33 period (according to Mishkin 1997) and which
in turn prompted the passing of the Glass-Steagall Act of 1933 and the establishment
of the Federal Deposit Insurance Corporation (FDIC) in 1934.
Although today banking panics are virtually nonexistent thanks to the FDIC and
related central banking policies, systemic risk exposures have taken shape in other
forms. In particular, the proliferation of hedge funds in recent years has indelibly
altered the risk/reward landscape of financial investments. Unregulated and opaque
investment partnerships that engage in a variety of active investment strategies,
hedge funds have generally yielded double-digit returns historically, but not without
commensurate risks, and such risks are currently not widely appreciated or well
understood. In particular, we argue that the risk/reward profile for most hedge funds
differs in important ways from more traditional investments, and such differences
may have potentially significant implications for systemic risk. One example is the
aftermath of the default of Russian government debt in August 1998, when LongTerm Capital Management (LTCM) and many other fixed-income hedge funds suffered catastrophic losses over the course of a few weeks, creating significant stress

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

on the global financial system and several major financial institutions—that is, creating systemic risk.
In this paper, we consider the impact of hedge funds on systemic risk by examining the unique risk-and-return profiles of hedge funds—at both the individual-fund
and the aggregate-industry level—and proposing some new risk measures for hedge
fund investments. Two major themes have emerged from August 1998: the importance of liquidity and leverage, and the capriciousness of correlations among instruments and portfolios that were thought to
The risk/reward profile for most hedge funds be uncorrelated. The precise mechanism
by which these two sets of issues posed
differs in important ways from more tradisystemic risks in 1998 is now well undertional investments, and such differences
stood. Because many hedge funds rely on
may have potentially significant implicaleverage, their positions are often considerably larger than the amount of collateral
tions for systemic risk.
posted to support those positions. Leverage
has the effect of a magnifying glass, expanding small profit opportunities into larger
ones but also expanding small losses into larger losses. And when adverse changes in
market prices reduce the market value of collateral, credit is withdrawn quickly, and the
subsequent forced liquidation of large positions over short periods of time can lead
to widespread financial panic, as in the aftermath of the default of Russian government debt in August 1998. The more illiquid the portfolio, the larger the price impact
of a forced liquidation, which erodes the fund’s risk capital that much more quickly.
Now if many funds face the same “death spiral” at a given point in time—that is, if
they become more highly correlated during times of distress and if those funds are
obligors of a small number of major financial institutions—then a market event like
August 1998 can cascade quickly into a global financial crisis. This is systemic risk.
Therefore, the two main themes of this study are illiquidity exposure and timevarying hedge fund correlations, both of which are intimately related to the dynamic
nature of hedge fund investment strategies and their risk exposures. In particular,
one of the justifications for the unusually rich fees that hedge funds charge is the fact
that highly skilled hedge fund managers are engaged in active portfolio management.
It is common wisdom that the most talented managers are drawn first to the hedge
fund industry because the absence of regulatory constraints enables them to make
the most of their investment acumen. With the freedom to trade as much or as little
as they like on any given day, to go long or short any number of securities and with
varying degrees of leverage, and to change investment strategies at a moment’s
notice, hedge fund managers enjoy enormous flexibility and discretion in pursuing
investment returns. But dynamic investment strategies imply dynamic risk exposures,
and while modern financial economics has much to say about the risk of static investments—the market beta is a sufficient statistic in this case—there is currently no single
summary measure of the risks of a dynamic investment strategy.1
To begin our discussion, we summarize the empirical properties of aggregate
and individual hedge fund data used in this study: the CSFB/Tremont hedge fund
indexes and the TASS individual hedge fund database. We then turn to the issue of
liquidity—one of the central aspects of systemic risk—and present several measures for gauging illiquidity exposure in hedge funds and other asset classes, which
we apply to individual and index data. Since systemic risk is directly related to
hedge fund failures, we investigate attrition rates of hedge funds in the TASS
database and present a logit analysis that yields estimates of a fund’s probability of
liquidation as a function of various fund characteristics such as return history,

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

assets under management (AUM), and recent fund flows. We then present estimates
of statistical regime-switching models for hedge fund indexes that capture certain
nonlinearities unique to the hedge fund industry. We conclude by discussing the
current industry outlook implied by the analytics and empirical results of this study.
Our tentative inferences suggest that the hedge fund industry may be heading into a
challenging period of lower expected returns and that systemic risk has been increasing steadily over the recent past. To address this growing concern, we put forward
a modest proposal to establish a new entity patterned after the U.S. National
Transportation Safety Board.
Our preliminary findings must be qualified by the acknowledgment that all of our
measures of systemic risk are indirect and therefore open to debate and interpretation. The main reason for this less-than-satisfying state of affairs is the fact that
hedge funds are currently not required to disclose any information about their risks
and returns to the public, so empirical studies of the hedge fund industry are based
only on very limited hedge fund data, provided voluntarily to TASS, and which may
or may not be representative of the industry as a whole. Even after February 1, 2006,
when, in response to the U.S. Securities and Exchange Commission’s (SEC’s) Rule
203(b)(3)–2 (which was subsequently struck down by the U.S. Court of Appeals in
June 2006), many hedge funds became registered investment advisers, the regular
filings of those funds did not include critical information such as a fund’s degree of
leverage, the liquidity of a fund’s portfolio, the identities of the fund’s major creditors
and obligors, and the specific terms under which the fund’s investors have committed their capital. Without this kind of information for the majority of funds in the
industry, it is virtually impossible, even for regulatory authorities like the SEC, to
construct direct measures of systemic risk. However, as the hedge fund industry
grows, the number and severity of hedge fund failures will undoubtedly increase as
well, eventually moving the industry toward greater transparency.

The Data
It is clear from our introduction that hedge funds exhibit unique and dynamic characteristics that bear further study. Fortunately, the returns of many individual hedge
funds are now available through a number of commercial databases such as AltVest,
CISDM, HedgeFund.net, HFR, and TASS. For the empirical analysis in this paper, we
use two main sources: (1) a set of aggregate hedge fund index returns from CSFB/
Tremont and (2) the TASS database of hedge funds, which consists of monthly
returns and accompanying information for 4,781 individual hedge funds (as of August
2004) from February 1977 to August 2004.2
The CSFB/Tremont indexes are asset-weighted indexes of funds with a minimum of
$10 million of AUM, a minimum one-year track record, and current audited financial
statements. An aggregate index is computed from this universe, and ten subindexes
based on investment style are also computed using a similar method. Indexes are
computed and rebalanced on a monthly frequency, and the universe of funds is redefined on a quarterly basis.

1. Accordingly, hedge fund track records are often summarized with multiple statistics, for example,
mean, standard deviation, Sharpe ratio, market beta, Sortino ratio, maximum drawdown, worst
month, etc.
2. For further information about these data see www.hedgeindex.com (CSFB/Tremont indexes) and
www.tremont.com (TASS). We also use data from Altvest, the University of Chicago’s Center for
Research in Security Prices, and Yahoo!Finance.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Table 1
Number of Funds in the TASS Hedge Fund Databases, February 1977–August 2004

Category
1
2
3
4
5
6
7
8
9
10
11

Number of TASS funds in
Graveyard

Definition

Live

Convertible arbitrage
Dedicated short bias
Emerging markets
Equity market neutral
Event driven
Fixed-income arbitrage
Global macro
Long/short equity
Managed futures
Multistrategy
Fund of funds

127
14
130
173
250
104
118
883
195
98
679

49
15
133
87
134
71
114
532
316
41
273

176
29
263
260
384
175
232
1,415
511
139
952

2,771

1,765

4,536

Total

Combined

The TASS database consists of monthly returns, AUM, and other fund-specific
information for 4,781 individual funds from February 1977 to August 2004. The
database is divided into two parts: “live” and “graveyard” funds. Hedge funds that are
in the “live” database are considered to be active as of August 31, 2004.3 As of August
2004, the combined database of both live and dead hedge funds contained 4,781
funds with at least one monthly return observation. Out of these 4,781 funds, 2,920
are in the live database and 1,861 in the graveyard database. The earliest data available for a fund in either database are from February 1977. TASS started tracking
dead funds in 1994; hence, it is only since 1994 that TASS transferred funds from the
live database to the graveyard database. Funds that were dropped from the live
database prior to 1994 are not included in the graveyard database, a circumstance
that may yield a certain degree of survivorship bias.4
The majority of 4,781 funds reported returns net of management and incentive
fees on a monthly basis.5 We eliminated 50 funds that reported only gross returns,
leaving 4,731 funds in the “combined” database (2,893 in the live and 1,838 in the
graveyard database). We also eliminated funds that reported returns on a quarterly—
not monthly—basis, leaving 4,705 funds in the combined database (2,884 in the live
and 1,821 in the graveyard database). Finally, we dropped funds that did not report
AUM, or reported only partial AUM, leaving a final sample of 4,536 hedge funds in the
combined database (2,771 funds in the live database and 1,765 funds in the graveyard database). For the empirical analysis in this paper, we impose an additional filter
in which we require funds to have at least five years of nonmissing returns, leaving
1,226 funds in the live database and 611 in the graveyard database for a combined
total of 1,837 funds. This filter obviously creates additional survivorship bias in the
remaining sample of funds, but since the main objective is to estimate measures of
illiquidity exposure and not to make inferences about overall performance, the filter
may not be as problematic. (See the studies cited in footnote 4.)
TASS also classifies funds into one of eleven different investment styles, listed in
Table 1 and described in the appendix, of which ten correspond exactly to the
CSFB/Tremont subindex definitions.6 Table 1 also reports the number of funds in

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

each category for the live, graveyard, and combined databases, and these numbers
show that the representation of investment styles is not evenly distributed but is concentrated among four categories: long/short equity (1,415), fund of funds (952), managed futures (511), and event driven (384). Together, these four categories account
for 71.9 percent of the funds in the combined database.
CSFB/Tremont indexes. Table 2 reports summary statistics for the monthly
returns of the CSFB/Tremont indexes from January 1994 to August 2004. Also included
for purposes of comparison are summary statistics for a number of aggregate measures
of market conditions.
Table 2 shows that there is considerable heterogeneity in the historical risk and
return characteristics of the various categories of hedge fund investment styles. For
example, the annualized mean return ranges from –0.69 percent for dedicated shortsellers to 13.85 percent for global macro, and the annualized volatility ranges from
3.05 percent for equity market neutral to 17.28 percent for emerging markets. The
correlations of the hedge fund indexes with the S&P 500 are generally low, with the
largest correlation at 57.2 percent for long/short equity and the lowest correlation at
–75.6 percent for dedicated short-sellers—as investors have discovered, hedge funds
offer greater diversification benefits than many traditional asset classes. However,
these correlations can vary over time. For example, consider a rolling sixty-month
correlation between the CSFB/Tremont Multi-Strategy Index and the S&P 500 from
January 1999 to December 2003, plotted in Figure 1. At the start of the sample in
January 1999, the correlation is –13.4 percent, then drops to –21.7 percent a year
later, and increases to 31.0 percent by December 2003 as the outliers surrounding
August 1998 drop out of the sixty-month rolling window.
Although changes in rolling correlation estimates are also partly attributable to
estimation errors,7 in this case, an additional explanation for the positive trend in
3. Once a hedge fund decides not to report its performance, is liquidated, is closed to new investment, restructured, or merged with other hedge funds, the fund is transferred into the graveyard
database. A hedge fund can only be listed in the graveyard database after being listed in the live
database. Because the TASS database fully represents returns and asset information for live and
dead funds, the effects of survivorship bias are minimized. However, the database is subject to
backfill bias; when a fund decides to be included in the database, TASS adds the fund to the live
database and includes all available prior performance of the fund. Hedge funds do not need to
meet any specific requirements to be included in the TASS database. Because of reporting delays
and time lags in contacting hedge funds, some graveyard funds can be incorrectly listed in the live
database for a period of time. However, TASS has adopted a policy of transferring funds from the
live to the graveyard database if they do not report over an eight- to ten-month period.
4. For studies attempting to quantify the degree and impact of survivorship bias, see Baquero, Horst, and
Verbeek (2005), Brown et al. (1992), Brown, Goetzmann, and Ibbotson (1999), Brown, Goetzmann,
and Park (2001), Carpenter and Lynch (1999), Fung and Hsieh (1997, 2000), Hendricks, Patel, and
Zeckhauser (1997), Horst, Nijman, and Verbeek (2001), and Schneeweis, Spurgin, and McCarthy (1996).
5. TASS defines returns as the change in net asset value during the month (assuming the reinvestment of any distributions on the reinvestment date used by the fund) divided by the net asset
value at the beginning of the month, net of management fees, incentive fees, and other fund
expenses. Therefore, these reported returns should approximate the returns realized by investors.
TASS also converts all foreign-currency-denominated returns to U.S.-dollar returns using the
appropriate exchange rates.
6. This correspondence is no coincidence—TASS is owned by Tremont Capital Management (acquired
by Lipper in March 2005), which created the CSFB/Tremont indexes in partnership with Credit
Suisse First Boston.
7. Under the null hypothesis of no correlation, the approximate standard error of the correlation
coefficient is 1/ √60 = 13%.

ECONOMIC REVIEW

Fourth Quarter 2006

ECONOMIC REVIEW

p-value
ρ3 of LB-Q

Fourth Quarter 2006

Sample
size

Annualized
mean

Annualized
SD

Corr. with
S&P 500

Min.

Med.

Max.

Skew.

Kurt.

ρ1

ρ2

CSFB/Tremont indexes
Hedge funds
Convertible arbitrage
Dedicated short-seller
Emerging markets
Equity market neutral
Event driven
Distressed
Event-driven multistrategy
Risk arbitrage
Fixed-income arbitrage
Global macro
Long/short equity
Managed futures
Multistrategy

128
128
128
128
128
128
128
128
128
128
128
128
128
125

10.51
9.55
–0.69
8.25
10.01
10.86
12.73
9.87
7.78
6.69
13.85
11.51
6.48
9.10

8.25
4.72
17.71
17.28
3.05
5.87
6.79
6.19
4.39
3.86
11.75
10.72
12.21
4.43

45.9
11.0
–75.6
47.2
39.6
54.3
53.5
46.6
44.7
–1.3
20.9
57.2
–22.6
5.6

–7.55
–4.68
–8.69
–23.03
–1.15
–11.77
–12.45
–11.52
–6.15
–6.96
–11.55
–11.43
–9.35
–4.76

0.78
1.09
–0.39
1.17
0.81
1.01
1.18
0.90
0.62
0.77
1.19
0.78
0.18
0.83

8.53
3.57
22.71
16.42
3.26
3.68
4.10
4.66
3.81
2.02
10.60
13.01
9.95
3.61

0.12
–1.47
0.90
–0.58
0.25
–3.49
–2.79
–2.70
–1.27
–3.27
0.00
0.26
0.07
–1.30

1.95
3.78
2.16
4.01
0.23
23.95
17.02
17.63
6.14
17.05
2.26
3.61
0.49
3.59

12.0
55.8
9.2
30.5
29.8
35.0
29.3
35.3
27.3
39.2
5.5
16.9
5.8
–0.9

4.0
41.1
–3.6
1.6
20.2
15.3
13.4
16.7
–1.9
8.2
4.0
6.0
–9.6
7.6

–0.5
14.4
0.9
–1.4
9.3
4.0
2.0
7.8
–9.7
2.0
8.8
–4.6
–0.7
18.0

54.8
0.0
73.1
0.7
0.0
0.0
0.3
0.0
1.2
0.0
65.0
21.3
64.5
17.2

S&P 500
Banks
LIBOR
USD
Oil
Gold
Lehman Bond
Large minus small cap
Value minus growth
Credit spread (not annualized)
Term spread (not annualized)
VIX (not annualized)

120
128
128
128
128
128
128
128
128
128
128
128

11.90
21.19
–0.14
–0.52
15.17
1.21
6.64
–1.97
0.86
4.35
1.65
0.03

15.84
13.03
0.78
7.51
31.69
12.51
4.11
13.77
18.62
1.36
1.16
3.98

100.0
55.8
3.5
7.3
–1.6
–7.2
0.8
7.6
–48.9
–30.6
–11.6
–67.3

–14.46
–18.62
–0.94
–5.35
–22.19
–9.31
–2.71
–20.82
–22.78
2.68
–0.07
–12.90

1.47
1.96
–0.01
–0.11
1.38
–0.17
0.50
0.02
0.40
3.98
1.20
0.03

9.78
11.39
0.63
5.58
36.59
16.85
3.50
12.82
15.85
8.23
3.85
19.48

–0.61
–1.16
–0.61
0.00
0.25
0.98
–0.04
–0.82
–0.44
0.82
0.42
0.72

0.30
5.91
4.11
0.08
1.17
3.07
0.05
5.51
3.01
–0.30
–1.25
4.81

–1.0
26.8
50.3
7.2
–8.1
–13.7
24.6
–13.5
8.6
94.1
97.2
–8.2

–2.2
6.5
32.9
–3.2
–13.6
–17.4
–6.3
4.7
10.2
87.9
94.0
–17.5

7.3
5.4
27.3
6.4
16.6
8.0
5.2
6.1
0.4
83.2
91.3
–13.9

86.4
1.6
0.0
71.5
7.3
6.2
3.2
36.6
50.3
0.0
0.0
5.8

Notes: The multistrategy return series begins in April 1994, and the S&P 500 return series ends in December 2003. “LB-Q” is the Ljung-Box (1978) Q-statistic.

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Table 2
Summary Statistics for Monthly CSFB/Tremont Hedge Fund Index Returns and
Various Hedge Fund Risk Factors, January 1994–August 2004

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Figure 1
Contemporaneous and Lagged Rolling Sixty-Month Correlation Between
CSFB/Tremont Multi-Strategy Index and S&P 500 Returns, January 1999–December 2003

40
30

Correlation (percent)

20
Lagged

10
0
Contemporaneous

–10
–20
–30
1999

2000

2001

2002

2003

2004

correlation is the enormous inflow of capital into multistrategy funds and fund of
funds over the past five years. As AUM increase, it becomes progressively more difficult for fund managers to implement strategies that are truly uncorrelated with
broad-based market indexes like the S&P 500. Moreover, Figure 1 shows that the correlation between the Multi-Strategy Index return and the lagged S&P 500 return has
also increased in the past year, indicating an increase in the illiquidity exposure of
this investment style (see Getmansky, Lo, and Makarov 2004 and the next section).
This increase in illiquidity exposure is also consistent with large inflows of capital
into the hedge fund sector.
Despite their heterogeneity, several indexes do share a common characteristic: negative skewness. Convertible arbitrage, emerging markets, event driven,
distressed, event-driven multistrategy, risk arbitrage, fixed-income arbitrage, and
multistrategy funds all have skewness coefficients less than zero, in some cases
substantially so. This property is an indication of tail risk exposure (see Lo 1999
for an explicit example involving short selling out-of-the-money put options on
the S&P 500 index) and is consistent with the nature of the investment strategies
employed by funds in those categories. For example, fixed-income arbitrage strategies are known to generate fairly consistent profits, with occasional losses that may
be extreme; hence, a skewness coefficient of –3.27 is not surprising. A more direct
measure of tail risk or “fat tails” is kurtosis; the normal distribution has a kurtosis
of 3.00, so values greater than this represent fatter tails than the normal. Not surprisingly, the two categories with the most negative skewness—event driven
(–3.49) and fixed-income arbitrage (–3.27)—also have the largest kurtosis, 23.95
and 17.05, respectively.
Several indexes also exhibit a high degree of positive serial correlation, as measured by the first three autocorrelation coefficients ρ1, ρ2, and ρ3, as well as the p-value
of the Ljung-Box Q-statistic, which measures the degree of statistical significance of

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

the first three autocorrelations.8 In comparison to the S&P 500, which has a firstorder autocorrelation coefficient of –1.0 percent, the autocorrelations of the hedge
fund indexes are very high, with values of 55.8 percent for convertible arbitrage, 39.2
percent for fixed-income arbitrage, and 35.0 percent for event driven, all of which are
significant at the 1 percent level according to the corresponding p-values. Serial correlation can be a symptom of illiquidity risk exposure, which is particularly relevant
for systemic risk, and we shall focus on this issue in more detail in the next section.
TASS data. Table 3 contains basic summary statistics for the funds in the TASS
live, graveyard, and combined databases. Not surprisingly, there is a great deal of
variation in mean returns and volatilities both across and within categories and
databases. For example, the 127 convertible arbitrage funds in the live database have
an average mean return of 9.92 percent and an average standard deviation of 5.51 percent, but in the graveyard database the forty-nine convertible arbitrage funds have
an average mean return of 10.02 percent and a much higher average standard deviation of 8.14 percent. Not surprisingly, average volatilities in the graveyard database
are uniformly higher than those in the live database because the higher-volatility funds
are more likely to be eliminated.9
Average serial correlations also vary considerably across categories in the combined database, but six categories stand out: convertible arbitrage (31.4 percent), fund
of funds (19.6 percent), event driven (18.4 percent), emerging markets (16.5 percent),
fixed-income arbitrage (16.2 percent), and multistrategy (14.7 percent). Given the
descriptions of these categories provided by TASS (see the appendix) and common
wisdom about the nature of the strategies involved—these categories include some of
the most illiquid securities traded—serial correlation seems to be a reasonable proxy
for illiquidity and smoothed returns (see Lo 2001; Getmansky, Lo, and Makarov 2004;
and the following section). Alternatively, equities and futures are among the most liquid
securities in which hedge funds invest, and not surprisingly, the average first-order
serial correlations for equity market neutral, long/short equity, and managed futures
are 5.1 percent, 9.5 percent, and –0.6 percent, respectively. Dedicated short-seller funds
also have a low average first-order autocorrelation, 5.9 percent, which is consistent
with the high degree of liquidity that often characterize short-sellers (by definition, the
ability to short a security implies a certain degree of liquidity).
These summary statistics suggest that illiquidity and smoothed returns may be
important attributes for hedge fund returns that can be captured to some degree by
serial correlation and the time-series model of smoothing discussed in the next section.

Measuring Illiquidity Risk
The different categories of hedge funds in the TASS database suggest that these funds
are likely to exhibit a heterogeneous array of risk exposures. However, a common
8. Ljung and Box (1978) propose the following statistic to measure the overall significance of the first
k autocorrelation coefficients: Q=T(T+2)Σ kj=1ρ^ 2j / (T–j), which is asymptotically χ 2k under the null
hypothesis of no autocorrelation. By forming the sum of squared autocorrelations, the statistic Q
^
reflects the absolute magnitudes of the ρ
js irrespective of their signs; hence, funds with large positive or negative autocorrelation coefficients will exhibit large Q-statistics. See Kendall, Stuart, and
Ord (1983, chap. 50.13) for further details.
9. This effect works at both ends of the return distribution—funds that are wildly successful are also
more likely to leave the database since they have less of a need to advertise their performance.
That the graveyard database also contains successful funds is supported by the fact that in some
categories, the average mean return in the graveyard database is the same as or higher than in the
live database—for example, convertible arbitrage, equity market neutral, and dedicated short-seller.

ECONOMIC REVIEW

Fourth Quarter 2006

Table 3
Means and Standard Deviations of Basic Summary Statistics for
Hedge Funds in the TASS Hedge Fund Databases, February 1977–August 2004

Category

Sample
size

Annualized
mean (%)
Mean
SD

Annualized
SD (%)
Mean
SD
Live funds
4.15
10.92
14.42
5.05
7.15
5.10
10.41
9.30
12.52
10.94
4.87

ρ1 (%)

Annualized
Sharpe ratio
Mean
SD

Ann. adjusted
Sharpe ratio
Mean
SD

Mean

33.6
3.5
18.8
4.4
19.4
16.4
1.3
11.3
3.4
18.5
22.9

19.2
10.9
13.8
22.7
20.9
23.6
17.1
17.9
13.9
21.3
18.5

2.57
–0.11
1.36
1.20
1.98
3.61
0.86
1.03
0.48
1.91
1.53

4.20
0.70
2.01
1.18
1.47
11.71
0.68
1.01
1.10
2.34
1.33

1.95
0.12
1.22
1.30
1.68
3.12
0.99
1.01
0.73
1.46
1.48

Ljung-Box

p-value (%)
Mean

2.86
0.46
1.40
1.28
1.47
7.27
0.79
0.95
0.63
2.06
1.16

19.5
48.0
35.5
41.6
31.3
36.6
46.8
38.1
52.3
31.1
33.7

27.1
25.7
31.5
32.6
34.1
35.2
30.6
31.8
30.8
31.7
31.6

Fourth Quarter 2006

127
14
130
173
250
104
118
883
195
98
679

9.92
0.33
17.74
6.60
12.52
9.30
10.51
13.05
8.59
12.65
6.89

5.89
11.11
13.77
5.89
8.99
5.61
11.55
10.56
18.55
17.93
5.45

5.51
25.10
21.69
7.25
8.00
6.27
13.57
14.98
19.14
9.31
6.14

Convertible arbitrage
Dedicated short-seller
Emerging markets
Equity market neutral
Event driven
Fixed-income arbitrage
Global macro
Long/short equity
Managed futures
Multistrategy
Fund of funds

49
15
133
87
134
71
114
532
316
41
273

10.02
1.77
2.74
7.61
9.07
5.51
3.74
9.69
4.78
5.32
4.53

6.61
9.41
27.74
26.37
15.04
12.93
28.83
22.75
23.17
23.46
10.07

8.14
27.54
27.18
12.35
12.35
10.78
21.02
23.08
20.88
17.55
13.56

Graveyard funds
6.08
25.5
18.79
8.1
18.96
14.3
13.68
6.4
12.10
16.6
9.97
15.9
18.94
3.2
16.82
6.4
19.35
–2.9
20.90
6.1
10.56
11.3

19.3
13.2
17.9
20.4
21.1
22.0
21.5
19.8
18.7
17.4
21.2

1.89
0.20
0.37
0.52
1.22
1.10
0.33
0.48
0.26
1.10
0.62

1.43
0.44
0.91
1.23
1.38
1.77
1.05
1.06
0.77
1.55
1.26

1.58
0.25
0.47
0.60
1.13
1.03
0.37
0.48
0.37
1.58
0.57

1.46
0.48
1.11
1.85
1.43
1.99
0.90
1.17
0.97
2.06
1.11

27.9
55.4
48.5
46.6
39.3
46.0
46.2
47.8
48.4
49.4
40.9

34.2
25.2
34.6
31.5
34.2
35.7
31.0
31.3
30.9
32.2
31.9

Convertible arbitrage
Dedicated short-seller
Emerging markets
Equity market neutral
Event driven
Fixed-income arbitrage
Global macro
Long/short equity
Managed futures
Multistrategy
Fund of funds

176
29
263
260
384
175
232
1,415
511
139
952

9.94
1.08
10.16
6.94
11.31
7.76
7.18
11.79
6.23
10.49
6.22

6.08
10.11
23.18
15.94
11.57
9.45
22.04
16.33
21.59
19.92
7.17

6.24
26.36
24.48
8.96
9.52
8.10
17.21
18.02
20.22
11.74
8.26

Combined funds
4.89
31.4
15.28
5.9
17.07
16.5
9.21
5.1
9.40
18.4
7.76
16.2
15.61
2.3
13.25
9.5
17.07
–0.6
15.00
14.7
7.75
19.6

19.5
12.2
16.2
21.9
21.0
22.9
19.3
18.8
17.4
20.9
20.0

2.38
0.05
0.86
0.97
1.71
2.59
0.60
0.82
0.34
1.67
1.27

3.66
0.59
1.63
1.24
1.48
9.16
0.92
1.06
0.91
2.16
1.37

1.85
0.19
0.84
1.06
1.49
2.29
0.70
0.81
0.50
1.49
1.21

2.55
0.46
1.31
1.53
1.48
5.86
0.90
1.07
0.88
2.05
1.22

21.8
52.0
42.2
43.3
34.1
40.4
46.5
41.7
49.8
36.7
35.8

29.3
25.2
33.7
32.3
34.3
35.6
30.8
31.9
30.9
32.9
31.8

Note: The p-values for the Ljung-Box (1978) Q-statistic for each fund use the first eleven autocorrelations of returns.

F E D E R A L R E S E R V E B A N K O F AT L A N TA

ECONOMIC REVIEW

Convertible arbitrage
Dedicated short-seller
Emerging markets
Equity market neutral
Event driven
Fixed-income arbitrage
Global macro
Long/short equity
Managed futures
Multistrategy
Fund of funds

F E D E R A L R E S E R V E B A N K O F AT L A N TA

theme surrounding systemic risk is credit and liquidity. Although they are separate
sources of risk exposures for hedge funds and their investors—one type of risk can
exist without the other—nevertheless, liquidity and credit have been inextricably
intertwined in the minds of most investors because of the problems encountered by
Long-Term Capital Management and many other fixed-income relative-value hedge
funds in August and September 1998. Because many hedge funds rely on leverage,
the size of the positions is often considerably larger than the amount of collateral supporting those positions. Leverage expands
While modern financial economics has
small profit opportunities into larger ones
but also expands small losses into larger
much to say about the risk of static investlosses. And when adverse changes in marments, there is currently no single sumket prices reduce collateral’s market value,
mary measure of the risks of a dynamic
credit is withdrawn quickly, and the subsequent forced liquidation of large positions
investment strategy.
over a short time can lead to widespread
financial panic, as occurred after the Russian government defaulted on its debt in
August 1998. Along with the many benefits of a truly global financial system is the
cost that a financial crisis in one country can have dramatic repercussions in several
others—that is, contagion.
The basic mechanisms driving liquidity and credit are familiar to most hedge fund
managers and investors, and the recent literature has made considerable progress in
modeling both credit and illiquidity risk. (See, for example, Bookstaber 1999, 2000
and Kao 2000 and their citations.) However, the complex network of creditor/obligor
relationships, revolving credit agreements, and other financial interconnections is
largely unmapped. Perhaps some of the newly developed techniques in the mathematical theory of networks will allow us to construct systemic measures for liquidity
and credit exposures and the robustness of the global financial system to idiosyncratic shocks. The “small-world” networks considered by Watts and Strogatz (1998)
and Watts (1999) seem to be particularly promising starting points.
A more immediate method for gauging the illiquidity risk exposure of a given
hedge fund is to examine the autocorrelation coefficients ρk of the fund’s monthly
returns, where ρk ≡ Cov[Rt, Rt–k]/Var[Rt] is the kth-order autocorrelation of {Rt},10 which
measures the degree of correlation between month t’s return and month t – k’s
return. To see why autocorrelations may be useful indicators of liquidity exposure,
recall that one of the earliest financial asset pricing models is the martingale model,
in which asset returns are serially uncorrelated (ρk = 0 for all k ≠ 0). Indeed, the title
of Samuelson’s (1965) seminal paper—“Proof that Properly Anticipated Prices
Fluctuate Randomly”—provides a succinct summary for the motivation of the martingale property: In an informationally efficient market, price changes must be
unforecastable if they are properly anticipated, that is, if they fully incorporate the
expectations and information of all market participants.
This extreme version of market efficiency is now recognized as an idealization
that is unlikely to hold in practice. (See, for example, Farmer and Lo 1999 and Lo
2004.) In particular, market frictions such as transactions costs, borrowing constraints, costs of gathering and processing information, and institutional restrictions
on short sales and other trading practices do exist, and they all contribute to the possibility of serial correlation in asset returns that cannot easily be “arbitraged” away
precisely because of the presence of these frictions. From this perspective, the
degree of serial correlation in an asset’s returns can be viewed as a proxy for the magnitude of the frictions, and illiquidity is one of most common forms of such frictions.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

For example, it is well known that the historical returns of residential real estate
investments are considerably more highly autocorrelated than, say, the returns of the
S&P 500 indexes during the same sample period. Similarly, the returns of S&P 500
futures contracts exhibit less serial correlation than those of the index itself. In both
examples, the more liquid instrument exhibits less serial correlation than the less liquid,
and the economic rationale is a modified version of Samuelson’s (1965) argument:
Predictability in asset returns will be exploited and eliminated only to the extent
allowed by market frictions. Despite the fact that the returns to residential real estate
are highly predictable, it is impossible to take full advantage of such predictability
because of the high transactions costs associated with real estate transactions, the
inability to short sell properties, and other frictions.11
A closely related phenomenon that buttresses this interpretation of serial correlation in hedge fund returns is the “nonsynchronous trading” effect, in which the
autocorrelation is induced in a security’s returns because those returns are computed
with closing prices that are not necessarily established at the same time each day (see,
for example, Campbell, Lo, and MacKinlay 1997, chap. 3). But in contrast to the studies
by Lo and MacKinlay (1988, 1990) and Kadlec and Patterson (1999), in which they
conclude that it is difficult to generate serial correlations in weekly U.S. equity portfolio returns much greater than 10 percent to 15 percent through nonsynchronous
trading effects alone, Getmansky, Lo, and Makarov (2004) argue that in the context
of hedge funds, significantly higher levels of serial correlation can be explained by the
combination of illiquidity and “performance smoothing” (see below), of which nonsynchronous trading is a special case. To see why, note that the empirical analysis in
the nonsynchronous-trading literature is devoted exclusively to exchange-traded
equity returns, not hedge fund returns; hence, the corresponding conclusions may
not be relevant in this context. For example, Lo and MacKinlay (1990) argue that
securities would have to go without trading for several days on average to induce
serial correlations of 30 percent, and they dismiss such nontrading intervals as unrealistic for most exchange-traded U.S. equity issues. However, such nontrading intervals are considerably more realistic for the types of securities held by many hedge
funds—for example, emerging-market debt, real estate, restricted securities, control
positions in publicly traded companies, asset-backed securities, and other exotic
over-the-counter derivatives. Therefore, nonsynchronous trading of this magnitude is
likely to be an explanation for the serial correlation observed in hedge fund returns.
But even when prices are synchronously measured—as they are for many funds
that mark their portfolios to market at the end of the month to strike a net asset value
at which investors can buy into or cash out of the fund—there are several other channels by which illiquidity exposure can induce serial correlation in the reported returns
of hedge funds. Apart from the nonsynchronous-trading effect, naive methods for
determining the fair market value or “marks” for illiquid securities can yield serially
correlated returns. For example, one approach to valuing illiquid securities is to
extrapolate linearly from the most recent transaction price (which, in the case of
10. The kth-order autocorrelation of a time series {Rt} is defined as the correlation coefficient
between Rt and Rt–k, which is simply the covariance between Rt and Rt–k divided by the square root
of the product of the variances of Rt and Rt–k. But since the variances of Rt and Rt–k are the same
under the assumption of stationarity, the denominator of the autocorrelation is simply the
variance of Rt.
11. These frictions have led to the creation of real-estate investment trusts (REITs), and the returns
to these securities—which are considerably more liquid than the underlying assets on which they
are based—exhibit much less serial correlation.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

emerging-market debt, might be several months ago), which yields a price path that
is a straight line, or at best a series of straight lines. Returns computed from such
marks will be smoother, exhibiting lower volatility and higher serial correlation than
true economic returns—that is, returns computed from mark-to-market prices where
the market is sufficiently active to allow all available information to be impounded
in the price of the security. Of course, for
Although they are separate sources of
securities that are more easily traded and
with deeper markets, mark-to-market prices
risk exposures for hedge funds and their
are more readily available, extrapolated
investors, liquidity and credit have been
marks are not necessary, and serial correinextricably intertwined in the minds of
lation is therefore less of an issue. But for
securities that are thinly traded, or not
most investors.
traded at all for extended periods of time,
marking them to market is often an expensive and time-consuming procedure that
cannot easily be performed frequently.12 Therefore, serial correlation may serve as a
proxy for a fund’s liquidity exposure.
Even if a hedge fund manager does not make use of any form of linear extrapolation
to mark the securities in his portfolio, he may still be subject to smoothed returns if
he obtains marks from broker-dealers that engage in such extrapolation. For example,
consider the case of a conscientious hedge fund manager attempting to obtain the
most accurate mark for his portfolio at month end by getting bid/offer quotes from
three independent broker-dealers for every security in his portfolio and then marking
each security at the average of the three quote midpoints. By averaging the quote
midpoints, the manager is inadvertently downward-biasing price volatility, and if any
of the broker-dealers employ linear extrapolation in formulating their quotes (and
many do, through sheer necessity because they have little else to go on for the most
illiquid securities), or if they fail to update their quotes because of light volume, serial
correlation will also be induced in reported returns.
Finally, a more prosaic channel by which serial correlation may arise in the reported
returns of hedge funds is through “performance smoothing,” the unsavory practice of
reporting only part of the gains in months when a fund has positive returns so as to
partially offset potential future losses and thereby reduce volatility and improve riskadjusted performance measures such as the Sharpe ratio. For funds containing liquid
securities that can be easily marked to market, performance smoothing is more difficult
and, as a result, less of a concern. Indeed, it is only for portfolios of illiquid securities
that managers and brokers have any discretion in marking their positions. Such practices are generally prohibited by various securities laws and accounting principles,
and great care must be exercised in interpreting smoothed returns as deliberate
attempts to manipulate performance statistics. After all, as discussed above, there are
many other sources of serial correlation in the presence of illiquidity, none of which
is motivated by deceit. Nevertheless, managers do have certain degrees of freedom
in valuing illiquid securities—for example, discretionary accruals for unregistered
private placements and venture capital investments—and Chandar and Bricker (2002)
conclude that managers of certain closed-end mutual funds do use accounting discretion to manage fund returns around a passive benchmark. Therefore, the possibility of deliberate performance smoothing in the less regulated hedge fund industry
must be kept in mind in interpreting any empirical analysis of serial correlation in
hedge fund returns.
Getmansky, Lo, and Makarov (2004) address these issues in more detail by first
examining other explanations of serial correlation in hedge fund returns that are

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

unrelated to illiquidity and smoothing—in particular, time-varying expected returns,
time-varying leverage, and incentive fees with high-water marks—and showing that
none of them can account for the magnitudes of serial correlation. They propose a
specific econometric model of smoothed returns that is consistent with both illiquidity exposure and performance smoothing, and they estimate it using the historical
returns of individual funds in the TASS hedge fund database. They find that funds
with the most significant amount of smoothing tend to be the more illiquid—for
example, emerging market debt, fixed-income arbitrage, etc.—and after correcting
for the effects of smoothed returns, some of the most successful types of funds tend
to have considerably less attractive performance characteristics.
However, for the purpose of developing a more highly aggregated measure to
address systemic risk exposure, a simpler approach is to use serial correlation coefficients and the Ljung-Box Q-statistic (see footnote 8). To illustrate this approach,
we estimate these quantities using monthly historical total returns of the ten largest
mutual funds (as of February 11, 2001) from various start dates through June 2000
and twelve hedge funds from various inception dates to December 2000. Monthly
total returns for the mutual funds were obtained from the University of Chicago’s
Center for Research in Securities Prices. The twelve hedge funds were selected from
the Altvest database to yield a diverse range of annual Sharpe ratios (from 1 to 5)
^
^
computed in the standard way (√12SR , where SR is the Sharpe ratio estimator
applied to monthly returns), with the additional requirement that the funds have a
minimum five-year history of returns.13 The names of the hedge funds have been
omitted to maintain their privacy, and we will refer to them only by their stated
investment styles, for example, relative value fund, risk arbitrage fund, etc.
ρ6, and the p-values of the
Table 4 reports the means, standard deviations, ^
ρ1 to ^
Q-statistic using the first six autocorrelations for the sample of mutual and hedge funds.
The first subpanel shows that the ten mutual funds have very little serial correlation
in returns, with first-order autocorrelations ranging from –3.99 percent to 12.37 percent, and with p-values of the corresponding Q-statistics ranging from 10.95 percent
to 80.96 percent, implying that none of the Q-statistics is significant at the 5 percent
level. The lack of serial correlation in these ten mutual fund returns is not surprising.
Because of their sheer size, these funds consist primarily of highly liquid securities,
and, as a result, their managers have very little discretion in marking such portfolios.
Moreover, many of the SEC regulations that govern the mutual fund industry—for
example, detailed prospectuses, daily net asset value calculations, and quarterly
filings—were enacted specifically to guard against arbitrary marks, price manipulation, and other unsavory investment practices.
The results for the twelve hedge funds are considerably different. In sharp contrast
to the mutual fund sample, the hedge fund sample displays substantial serial correlation, with first-order autocorrelation coefficients that range from –20.17 percent to
49.01 percent, with eight out of twelve funds that have Q-statistics with p-values less
than 5 percent and ten out of twelve funds with p-values less than 10 percent. The
only two funds with p-values that are not significant at the 5 percent or 10 percent
levels are the risk arbitrage A and risk arbitrage B funds, which have p-values of
74.10 percent and 93.42 percent, respectively. These results are consistent with the
notion of serial correlation as a proxy for illiquidity risk because among the various
12. Liang (2003) presents a sobering analysis of the accuracy of hedge fund returns that underscores
the challenges of marking a portfolio to market.
13. See www.investorforce.com for further information about the Altvest database.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Table 4
Summary Statistics for Monthly Total Returns of Mutual Funds and Hedge Funds
Fund

Start

Sample
Size

µ^
(%)

σ^
(%)

ρ1
(%)
^

ρ2
(%)
^

ρ3
(%)

ρ4
(%)

ρ5
(%)

ρ6
(%)

p(Q6)
(%)

Mutual funds
Vanguard 500 Index

76.10

286

1.30

4.27

–4.0 –6.6

–4.9

–6.4

10.1

–3.6

31.9

Fidelity Magellan

67.01

402

1.73

6.23

12.4 –2.3

–0.4

0.7

7.1

3.1

17.8

Investment Company of America

63.01

450

1.17

4.01

Janus

70.03

364

1.52

4.75

Fidelity Contrafund

67.05

397

1.29

Washington Mutual Investors

63.01

450

1.13

Janus Worldwide

92.01

102

Fidelity Growth and Income

86.01

American Century Ultra

81.12

Growth Fund of America

1.8 –3.2

–4.5

–1.6

6.3

–5.6

55.9

0.0

–3.7

–8.2

2.1

–0.6

30.3

4.97

7.4 –2.5

–6.8

–3.9

2.7

–4.5

42.3

4.09

–0.1 –7.2

–2.6

0.7

11.6

–2.6

16.7

1.81

4.36

11.4

–3.8 –15.4 –21.4 –10.3

11.0

174

1.54

4.13

5.1 –1.6

–8.2 –15.6

2.1

–7.3

30.9

223

1.72

7.11

2.3

3.4

1.4

–3.7

–7.9

–6.0

81.0

64.07

431

1.18

5.35

8.5 –2.7

–4.1

–3.2

3.4

0.3

52.5

Convertible/option arbitrage

92.05

104

1.63

0.97

42.6 29.0

21.4

2.9

–5.9

–9.7

0.0

Relative value

92.12

0.66

0.21

25.9 19.2

–2.1 –16.4

–6.2

1.4

3.3

10.5

3.4

Hedge funds

Mortgage-backed securities

93.01

1.33

0.79

42.0 22.1

16.7

22.6

6.6

–2.0

0.0

High-yield debt

94.06

1.30

0.87

33.7 21.8

13.1

–0.8

13.8

4.0

1.1

Risk arbitrage A

93.07

1.06

0.69

–4.9 –10.8

6.9

–8.5

9.9

3.1

74.1

Long/short equities

89.07

138

1.18

0.83 –20.2 24.6

8.7

11.2

13.5

16.9

0.1

Multistrategy A

95.01

1.08

0.75

48.9 23.4

3.4

0.8

–2.3 –12.8

0.1

Risk arbitrage B

94.11

0.90

0.77

–4.9

–8.3

–5.7

Convertible arbitrage A

92.09

100

1.38

1.60

33.8 30.8

7.9

–9.4

Convertible arbitrage B

94.07

0.78

0.62

32.4

9.7

–4.5

6.5

Multistrategy B

89.06

139

1.34

1.63

49.0 24.6

10.6

Fund of funds

94.10

1.68

2.29

29.7 21.2

0.9

2.5

8.9

0.6

9.8

93.4

3.6

–4.4

0.1

–6.3 –10.6

8.6

7.8

7.5

0.0

–0.9 –12.4

3.0

6.8

Notes: Figures reflect various start dates through June 2000 for the mutual fund sample and through December 2000 for the hedge fund sam^ ” denotes the kth autocorrelation coefficient, and “p (Q )” denotes the significance level of the Ljung-Box (1978) Q-statistic,
ple. “ρ
6
k
T (T+2)Σ6k–1 ρ2k /T– k ), which is asymptotically χ 26 under the null hypothesis of no serial correlation.
Source: AlphaSimplex Group

types of funds in this sample, risk arbitrage is likely to be the most liquid, since, by
definition, such funds invest in securities that are exchange traded and where trading volume is typically heavier than usual because of the impending merger events on
which risk arbitrage is based.
Having established the relevance of serial correlation as a proxy for illiquidity, we
now turn to the measurement of illiquidity in the context of systemic risk. To that
end, let ρ1t,i denote the first-order autocorrelation coefficient in month t for fund i
using a rolling window of past returns. Then an aggregate measure of illiquidity ρ*t in
the hedge fund sector may be obtained by a cross-sectional weighted average of
these rolling autocorrelations, where the weights ωit are simply the proportion of
AUM for fund i:
Nt

∗
(1) ρt ≡ ∑ ω it ρ1t ,i ,
i=1

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

(2) ω it ≡

AUM it ,
∑ AUM jt
Nt
j =1

where Nt is the number of funds in the sample in month t and AUMjt is the AUM for
fund j in month t.
Figure 2 plots these weighted correlations from January 1980 to August 2004
using all funds in the TASS combined database with at least thirty-six consecutive
trailing months of nonmissing returns, along with the number of funds each month
(at the bottom, measured by the right vertical axis), and the median correlation in
the cross section.14 The median correlation is quite different from the asset-weighted
correlation in the earlier part of the sample, but as the number of funds increases
over time, the behavior of the median becomes closer to that of ρ*t .
Figure 2 also shows considerable swings in ρ*t over time, with dynamics that seem
to be related to liquidity events. In particular, consider the following events: Between
November 1980 and July 1982 the S&P 500 dropped 23.8 percent. In October 1987
the S&P 500 fell by 21.8 percent. In 1990 the Japanese “bubble economy” burst. In
August 1990 the Persian Gulf War began with Iraq’s invasion of Kuwait, ending in
January 1991 with Kuwait’s liberation by coalition forces. In February 1994 the U.S.
Federal Reserve started a tightening cycle that caught many hedge funds by surprise,
causing significant dislocation in bond markets worldwide. The end of 1994 witnessed
the start of the “Tequila crisis” in Mexico. In August 1998 Russia defaulted on its government debt. And between August 2000 and September 2002 the S&P 500 fell by
46.3 percent. In each of these cases, the weighted autocorrelation rose in the aftermath, and in most cases abruptly. Of course, the fact that we are using a thirty-sixmonth rolling window suggests that as outliers drop out of the window, correlations
can shift dramatically. However, as a coarse measure of liquidity in the hedge fund
sector, the weighted autocorrelation seems to be intuitively appealing and informative. Figure 2 shows that over the most recent past, the weighted autocorrelation is
on the rise, implying that hedge funds are taking more illiquidity exposure. This is
another indirect indicator of a rise in systemic risk in the hedge fund industry.

Hedge Fund Liquidations
Since the collapse of LTCM in 1998, it has become clear that hedge fund liquidations
can be a significant source of systemic risk. In this section, we consider several measures of liquidation probabilities for hedge funds in the TASS database, including a
review of hedge fund attrition rates documented in Getmansky, Lo, and Mei (2004)
and a logit analysis of hedge funds liquidations in the TASS graveyard database. By
analyzing the factors driving hedge fund liquidations, we may develop a broader
understanding of the likely triggers of systemic risk in this sector.
Because of the voluntary nature of inclusion in the TASS database, graveyard funds
do not consist solely of liquidations. TASS gives one of seven distinct reasons for each
fund that is assigned to the graveyard, ranging from “liquidated” (status code 1) to
“unknown” (status code 9). It may seem reasonable to confine our attention to those
graveyard funds categorized as liquidated or perhaps to drop those funds that are
closed to new investment (status code 4) from our sample. However, because our
purpose is to develop a broader perspective on the dynamics of the hedge fund industry, we argue that using the entire graveyard database may be more informative. For
14. The number of funds in the early years is relatively low, reaching a level of fifty or more only in
late 1988; therefore, the weighted correlations before then may be somewhat less informative.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Figure 2
Mean and Median Autocorrelations of Hedge Funds in the
TASS Combined Database, January 1980–August 2004
40

5,000
4,500

Asset-weighted autocorrelation

4,000
3,500
3,000

2,500
0

2,000

Number of funds

Autocorrelation (percent)

1,500

Number of funds

1,000

–10

500

Median autocorrelation

–20

0
1980

1984

1988

1992

1996

2000

2004

Note: Monthly cross-sectional median and weighted-mean first-order autocorrelation coefficients are for individual hedge funds in the TASS combined database with at least thirty-six trailing months of returns.

example, by eliminating graveyard funds that are closed to new investors, we create
a downward bias in the performance statistics of the remaining funds. Because we do
not have detailed information about each of these funds, we cannot easily determine
how any particular selection criterion will affect the statistical properties of the
remainder. Therefore, we choose to include the entire set of graveyard funds in our
analysis but caution readers to keep in mind the composition of this sample when
interpreting our empirical results.
To estimate the influence of various hedge fund characteristics on the likelihood
of liquidation, in this section we report the results of a logit analysis of liquidations in
the TASS database. Logit can be viewed as a generalization of the linear regression
model to situations in which the dependent variable takes on only a finite number of
discrete values (see, for example, Maddala 1983 for details). To estimate the logit model
of liquidation, we use a sample of 4,536 TASS funds from February 1977 to August
2004, of which 1,765 are in the graveyard database and 2,771 are in the live database.
As discussed earlier, the graveyard database was initiated only in January 1994; hence,
this will be the start date of our sample for purposes of estimating the logit model of
liquidation. For tractability, we focus on annual observations only, so the dependent
variable Zit indicates whether fund i is live or liquidated in year t.15 Over the sample
period from January 1994 to August 2004, we have 23,925 distinct observations for
Zit, and after filtering out funds that do not have at least two years of history, we are
left with 12,895 observations.
Associated with each Zit is a set of explanatory variables listed in Table 5. The
motivation for AGE, ASSETS, and RETURN is well known—older funds, funds with
greater assets, and funds with better recent performance are all less likely to be liquidated, so we would expect negative coefficients for these explanatory variables
(recall that a larger conditional mean for Z * implies a higher probability that Zit = 1 or
64

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Table 5
Definition of Explanatory Variables in Logit Analysis of Annual
Hedge Fund Liquidations in the TASS Database, January 1994–August 2004
Variable

Definition

AGE

The current age of the fund (in months)

ASSETS

The natural logarithm of current total assets under management

ASSETS–1

The natural logarithm of total assets under management as of December 31 of the previous year

RETURN

Current year-to-date total return

RETURN–1

Total return last year

RETURN–2

Total return two years ago

FLOW

Fund’s current year-to-date total dollar inflow divided by previous year’s assets under management,
where dollar inflow in month τ is defined as FLOWτ = AUM τ – AUM τ–1 (1 + R τ) and AUM τ is the
total assets under management at the beginning of month τ, Rτ is the fund’s net return for month τ,
and year-to-date total dollar inflow is simply the cumulative sum of monthly inflows since January
of the current year

FLOW–1

Previous year’s total dollar inflow divided by assets under management the year before

FLOW–2

Total dollar inflow two years ago divided by assets under management the year before

liquidation). The FLOW variable is motivated by the well-known “return-chasing”
phenomenon in which investors flock to funds that have had good recent performance
and leave funds that have underperformed (see, for example, Chevalier and Ellison
1997; Sirri and Tufano 1998; and Agarwal, Daniel, and Naik 2004). Because AUM is
highly persistent—with a correlation of 94.3 percent between its contemporaneous
and lagged values—we include only the lagged variable ASSETS–1 in our logit analysis,
yielding the following specification, which we call model 1:
(3) Zit = G(β0 + β1AGEit + β2ASSETSit–1 +
β3RETURNit + β4RETURNit–1 + β5RETURNit–2 +
β6FLOWit + β7FLOWit–1 +β8FLOWit–2 + εit).
Table 6 contains maximum-likelihood estimates of equation (3) in the first three
columns, with statistically significant parameters in bold. Note that most of the parameter estimates are highly significant. This significance results from the unusually large
sample size, which typically yields statistically significant estimates because of the
small standard errors implied by large samples (recall that the standard errors of
consistent and asymptotically normal estimators converge to 0 at a rate of 1/ n
where n is the sample size). This result suggests that we may wish to impose a higher

15. Note that a fund cannot “die” more than once, so liquidation occurs exactly once for each fund i in
the graveyard database. In particular, the time series observations of funds in this database will
always be {0, 0,…, 0, 1}. This fact suggests that a more appropriate statistical technique for modeling hedge fund liquidations is survival analysis, which we plan to pursue in a future study.
However, for purposes of summarizing the impact of certain explanatory variables on the probability of hedge fund liquidations, logit analysis is a reasonable choice.

ECONOMIC REVIEW

Fourth Quarter 2006

ECONOMIC REVIEW

Model 1

Model 2

p-value
Variable

Sample Size
R2 (%)

Fourth Quarter 2006

Constant
AGE
ASSETS–1
RETURN
RETURN–1
RETURN–2
FLOW
FLOW–1
FLOW–2
I(1994)
I(1995)
I(1996)
I(1997)
I(1998)
I(1999)
I(2000)
I(2001)
I(2002)
I(2003)
I(ConvertArb)
I(DedShort)
I(EmrgMkt)
I(EqMktNeut)
I(EventDr)
I(FixedInc)
I(GlobMac)
I(LongShortEq)
I(MgFut)
I(Multistrat)

SE(ββ)

(%)

12,895
29.3
4.73
–0.03
–0.26
–2.81
–1.39
–0.04
–0.63
–0.13
–0.09

0.34
0.00
0.02
0.19
0.16
0.09
0.08
0.04
0.02

Model 3

p-value
SE(ββ)

(%)

12,895
34.2
<.01
<.01
<.01
<.01
<.01
67.5
<.01
0.0
<.01

2.31
–0.03
–0.19
–2.86
–1.40
–0.38
–0.49
–0.11
–0.11
0.79
1.24
1.83
1.53
1.81
2.10
2.25
1.97
1.46
1.55
0.44
0.05
0.25
0.12
0.33
0.50
0.32
0.18
0.49
0.17

0.41
0.00
0.02
0.20
0.17
0.14
0.07
0.03
0.02
0.38
0.27
0.20
0.21
0.18
0.18
0.17
0.17
0.16
0.16
0.20
0.37
0.15
0.20
0.15
0.19
0.18
0.11
0.12
0.25

Model 4

p-value
SE(ββ)

(%)

12,895
34.2
<.01
<.01
<.01
<.01
<.01
0.7
<.01
0.1
<.01
3.9
<.01
<.01
<.01
<.01
<.01
<.01
<.01
<.01
<.01
2.9
88.9
10.2
54.7
3.0
1.1
7.4
10.2
<.01
49.4

–5.62
–1.62
–0.34
–0.67
–0.36
–0.12
–32.72
–7.53
–1.74
0.79
1.24
1.83
1.53
1.81
2.10
2.25
1.97
1.46
1.55
0.44
0.05
0.25
0.12
0.33
0.50
0.32
0.18
0.49
0.17

Note: The dependent variable Z takes on the value 1 in the year a hedge fund is liquidated and is 0 in all prior years.

0.18
0.07
0.04
0.05
0.04
0.04
4.91
2.33
0.36
0.38
0.27
0.20
0.21
0.18
0.18
0.17
0.17
0.16
0.16
0.20
0.37
0.15
0.20
0.15
0.19
0.18
0.11
0.12
0.25

Model 5

p-value
SE(ββ)

(%)

p-value
β

12,846
34.5
<.01
<.01
<.01
<.01
<.01
0.7
<.01
0.1
<.01
3.9
<.01
<.01
<.01
<.01
<.01
<.01
<.01
<.01
<.01
2.9
88.9
10.2
54.7
3.0
1.1
7.4
10.2
<.01
49.4

–5.67
–1.66
–0.36
–0.67
–0.36
–0.12
–33.27
–7.60
–1.64
0.82
1.18
1.83
1.52
1.80
2.05
2.19
1.96
1.41
1.53
0.43
–0.03
0.24
0.15
0.31
0.45
0.24
0.15
0.49
0.18

0.18
0.07
0.04
0.05
0.04
0.05
5.04
2.37
0.36
0.39
0.28
0.21
0.21
0.19
0.18
0.17
0.17
0.16
0.16
0.20
0.39
0.15
0.20
0.15
0.20
0.18
0.11
0.13
0.25

SE(ββ)

(%)

12,310
35.4
<.01
<.01
<.01
<.01
<.01
1.1
<.01
0.1
<.01
3.4
<.01
<.01
<.01
<.01
<.01
<.01
<.01
<.01
<.01
3.4
94.3
11.7
46.7
4.7
2.3
20.2
16.6
0.0
48.5

–7.04
–2.08
–0.38
–0.61
–0.44
–0.17
–32.93
–19.26
–1.83
1.01
1.37
1.92
2.03
2.29
2.25
2.08
1.80
1.50
1.71
0.16
0.20
0.54
0.53
–0.01
0.33
0.33
0.14
0.71
0.85

0.26
0.10
0.06
0.06
0.06
0.07
6.74
4.71
0.51
0.54
0.37
0.28
0.27
0.24
0.24
0.24
0.25
0.22
0.22
0.34
0.49
0.20
0.25
0.24
0.30
0.25
0.15
0.16
0.29

<.01
<.01
<.01
<.01
<.01
1.3
<.01
<.01
0.0
5.9
0.0
<.01
<.01
<.01
<.01
<.01
<.01
<.01
<.01
62.5
68.0
0.7
3.4
97.4
26.8
17.9
36.4
<.01
0.3

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Table 6
Maximum Likelihood Estimates of a Logit Model for Annual Hedge Fund Liquidations
in the TASS Database, January 1994–August 2004

F E D E R A L R E S E R V E B A N K O F AT L A N TA

threshold of statistical significance in this case, so as to provide a better balance
between type I and type II errors.16
The negative signs of all the coefficients other than the constant term confirm our
intuition that age, AUM, cumulative return, and fund flows all have a negative impact
on the probability of liquidation. The fact that RETURN–2 is not statistically significant suggests that the most recent returns have the highest degree of relevance for
hedge fund liquidations, a possible indication of the short-term performance-driven
nature of the hedge fund industry. The R2 of this regression is 29.3 percent, which
implies a reasonable level of explanatory power for this simple specification.17
To address fixed effects associated with the calendar year and hedge fund style
category, in model 2 we include indicator variables for ten out of eleven calendar years
and ten out of eleven hedge fund categories, yielding the following specification:
10

k=1

(4) Zit = G[β0 + ∑ ζ kI( YEAR k,i,t ) + ∑ ξ kI( CAT k,i,t ) +
β1AGEit + β2ASSETSit–1 +
β3RETURNit + β4RETURNit–1 + β5RETURNit–2 +
β6FLOWit + β7FLOWit–1 +β8FLOWit–2 + εit],
where

⎧1 if t = k
;
(5a) I ( YEAR k,i,t ) ≡ ⎨
⎩0 otherwise
⎧1 if fund i isin category k
(5b) I ( CATk,i,t ) ≡ ⎨
.
⎩0 otherwise

The columns labelled “model 2” in Table 6 contain the maximum-likelihood estimates
of (4) for the same sample of funds as model 1. The coefficients for AGE, ASSETS, and
RETURN exhibit the same qualitative properties as in model 1, but the fixed-effect
variables do provide some additional explanatory power, yielding an R2 of 34.2 percent. In particular, the coefficients for the 1999 and 2000 indicator variables are higher
than those of the other year indicators, a manifestation of the impact of August 1998
and the collapse of LTCM and other fixed-income relative-value hedge funds. The
impact of LTCM can also be seen from the coefficients of the category indicators—at
0.50, fixed-income relative value has the largest estimate among all ten categories.
The managed futures category has a comparable coefficient of 0.49, which is consistent with the higher volatility of such funds and the fact that this category exhibits
the highest attrition rate, 14.4 percent, during the 1994–2003 sample period (see
Getmansky, Lo, and Mei 2004 for a more detailed discussion of hedge fund attrition
rates). However, the fact that the convertible arbitrage and event driven categories
are the next largest, with coefficients of 0.44 and 0.33, respectively, is somewhat surprising given their unusually low attrition rates of 5.2 percent and 5.4 percent,
respectively (see Getmansky, Lo, and Mei 2004). This fact suggests that the conditional probabilities produced by a logit analysis—which control for AUM, fund flows,
and performance—yields information not readily available from the unconditional
16. See Leamer (1978) for further discussion of this phenomenon, known as “Lindley’s paradox.”
17. This R 2 is the adjusted generalized coefficient of determination proposed by Nagelkerke (1991),
which renormalizes Cox and Snell’s (1989) R 2 measure by its maximum (which is less than unity)
so that it spans the entire unit interval. See Nagelkerke (1991) for further discussion.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

frequency counts of simple attrition statistics. The remaining category indicators are
statistically insignificant at the 5 percent level.
To facilitate comparisons across explanatory variables, we standardize each of the
nonindicator explanatory variables by subtracting its mean and dividing by its standard
deviation and then reestimating the parameters of equation (4) via maximum likelihood. This procedure yields estimates that are renormalized to standard deviation units
of each explanatory variable and are contained in the columns labelled “model 3” of
Table 6. The renormalized estimates show that fund flows are an order of magnitude
more important in determining the probability of liquidation than AUM, returns, or age,
with normalized coefficients of –32.72 and –7.53 for FLOW and FLOW–1, respectively.
Finally, we reestimate the logit model in equation (4) for two subsets of funds
using standardized explanatory variables. In model 4, we omit graveyard funds that
have either merged with other funds or are closed to new investments (status codes
4 and 5), yielding a subsample of 12,846 observations. And in model 5, we omit all
graveyard funds except those that have liquidated (status code 1), yielding a subsample of 12,310 observations. The last two sets of columns in Table 6 show that the
qualitative features of most of the estimates are unchanged, with the funds in model
5 exhibiting somewhat higher sensitivity to the lagged FLOW variable. However, the
category fixed effects in model 5 do differ in some ways from those of models 2–4,
with significant coefficients for emerging markets, equity market neutral, and multistrategy, as well as for managed futures, suggesting significant differences between
the full graveyard sample and the subsample of funds with status code 1.
Because of the inherent nonlinearity of the logit model, the coefficients of the
explanatory variables cannot be as easily interpreted as in the linear regression
model. One way to remedy this situation is to compute the estimated probability of
^
liquidation implied by the parameter estimates β and specific values for the explanatory variables, which is readily accomplished by observing that

(

)

(

(6a) pit ≡ Prob Zit = 1 = Prob Zit* > 0

(6b)

=Prob( Χ′itβ + ε it > 0) =

(6c) pˆ it =

(

)

exp Χ′itβ

(

)

1 + exp Χ′itβ

)

;

exp( Χ′itβˆ )
.
1 + exp( Χ′ βˆ )
it

Table 7 reports year-by-year summary statistics for the estimated liquidation
^
^
probabilities {ρ
it} of each fund in our sample, where each ρit is computed using values
of the explanatory variables in year t. The left panel of Table 7 contains summary
statistics for estimated liquidation probabilities from model 1, and the right panel
contains corresponding figures from model 5. We have also stratified the estimated
liquidation probabilities by their liquidation status—live funds, graveyard funds, and
the combined sample.18
For both models 1 and 5, the mean and median liquidation probabilities are higher
for graveyard funds than for live funds, a reassuring sign that the explanatory variables are indeed providing explanatory power for the liquidation process. For model 1,
the combined sample shows an increase in the mean and median liquidation probabilities in 1998, as expected, and another increase in 2001, presumably due to the
bursting of the technology bubble in U.S. equity markets. Most troubling from the

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

perspective of systemic risk, however, is the fact that the mean and median liquidation probabilities for 2004 (which includes data only up to August) are 11.24 percent
and 7.69 percent, respectively, the highest levels in our entire sample. This result
may be a symptom of the enormous growth that the hedge fund industry has enjoyed
in recent years, which increases both the number of funds entering and exiting the
industry, but may also indicate more challenging market conditions for hedge funds
in the coming months. Note that the mean and median liquidation probabilities for
model 5 do not show the same increase in 2004—another manifestation of the time
lag with which the graveyard database is updated. (Recall that model 5 includes only
those funds with status code 1, but a large number of funds that eventually receive
this classification have not yet reached their eight- to ten-month limit by August
2004.) Therefore, model 1’s estimated liquidation probabilities are likely to be more
accurate for the current year.19
The logit estimates and implied probabilities suggest that a number of factors
influence the likelihood of a hedge fund’s liquidation, including past performance,
AUM, fund flows, and age. Given these factors, our estimates imply that the average liquidation probability for funds in 2004 is over 11 percent, which is higher than
the historical unconditional attrition rate of 8.8 percent. To the extent that a series
of correlated liquidations stresses the capital reserves of financial counterparties,
this is yet another indirect measure of an increase in systemic risk from the hedge
fund industry.

Regime-Switching Models
Our final hedge fund-based measure of systemic risk is motivated by the phaselocking example of Lo (1999), where the return-generating process exhibits apparent changes in expected returns and volatility that are discrete and sudden—for
example, the Mexican peso crisis of 1994–95, the Asian crisis of 1997, and the global
flight to quality precipitated by the default of Russian GKO debt in August 1998.
Linear models are generally incapable of capturing such discrete shifts; hence, more
sophisticated methods are required. In particular, we propose to model such shifts by
a regime-switching process in which two states of the world are hypothesized, and
the data are allowed to determine the parameters of these states and the likelihood
of transitioning from one to the other. Regime-switching models have been used in a
number of contexts, ranging from Hamilton’s (1989) model of the business cycle to
Ang and Bekaert’s (2004) regime-switching asset allocation model, and we propose
to apply it to the CSFB/Tremont indexes to obtain another measure of systemic risk,
that is, the possibility of switching from a normal to a distressed regime.
Denote by Rt the return of a hedge fund index in period t and suppose Rt satisfies
the following:
(7a) Rt = It · R1t + (1 – It) · R2t;

(

)

(7b) Rit ~ N ui ,σ 2i ;
18. Note that the usage of “graveyard funds” in this context is somewhat different, involving a time
dimension as well as liquidation status. For example, in this context the set of graveyard funds in
1999 refers to only those funds that liquidated in 1999 and does not include liquidations before
or after 1999.
19. The TASS reporting delay also affects model 1, suggesting that its estimated liquidation probabilities for 2004 are biased downward as well.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Table 7
Liquidation Probabilities of Logit Models for
Annual Hedge Fund Liquidations, January 1994–August 2004
1994

1995

1996

1997

Model 1
1998
1999

2000

2001

2002

2003

2004

Mean
SD
Min
10%
25%
50%
75%
90%
Max
Count

4.19
7.49
0.01
0.13
0.43
1.16
4.21
12.13
52.49
357

5.47
9.33
0.01
0.19
0.51
1.46
6.03
16.17
58.30
483

5.84
11.15
0.00
0.19
0.52
1.52
5.11
16.85
72.97
629

5.04
9.74
0.00
0.18
0.56
1.59
4.83
13.27
90.06
773

Live funds
6.32
5.17
9.66
8.61
0.00
0.00
0.31
0.20
0.99
0.79
2.71
2.18
7.20
5.55
16.76 12.80
77.63 87.06
924 1,083

5.59
8.15
0.00
0.35
1.10
2.80
6.54
13.78
75.83
1,207

6.84
9.23
0.00
0.44
1.39
3.69
8.39
16.23
92.36
1,317

8.92
10.15
0.00
0.68
2.05
5.62
12.01
21.61
79.02
1,480

7.11
8.00
0.00
0.41
1.45
4.49
10.22
17.26
92.44
1,595

11.04
10.91
0.00
0.89
2.66
7.55
16.31
26.33
79.96
1,898

Mean
SD
Min
10%
25%
50%
75%
90%
Max
Count

36.59
24.46
4.91
6.08
22.06
32.82
48.40
71.63
77.37
10

32.85
22.77
2.50
8.39
16.28
28.53
49.79
58.62
97.42
27

31.89
18.86
1.05
10.63
17.47
27.44
43.36
60.08
79.51
73

39.75
22.70
0.25
9.29
21.81
39.78
56.94
71.13
88.70
62

Graveyard funds
30.64 27.68
21.67 19.24
0.00
0.53
6.86
4.98
12.13 12.84
25.20 24.03
46.21 39.62
61.74 50.75
85.41 84.87
104
129

22.78
17.67
0.22
2.41
9.14
19.81
34.92
45.84
87.89
176

28.17
20.03
0.98
5.94
12.07
23.28
41.01
58.90
78.68
175

25.22
18.22
0.13
5.50
10.58
21.50
37.98
48.81
94.65
167

21.55
15.91
0.02
2.64
8.32
19.18
32.28
45.42
72.29
158

17.01
14.30
0.25
2.26
6.43
13.35
25.26
34.67
67.10
68

Mean
SD
Min
10%
25%
50%
75%
90%
Max
Count

5.07
9.86
0.01
0.14
0.45
1.23
4.89
14.96
77.37
367

6.92
12.10
0.01
0.20
0.55
1.72
7.67
20.53
97.42
510

8.55
14.53
0.00
0.22
0.62
1.84
8.96
27.36
79.51
702

7.61
14.44
0.00
0.20
0.62
1.88
6.25
22.94
90.06
835

Combined funds
8.78
7.56
13.59 12.39
0.00
0.00
0.38
0.22
1.10
0.91
3.34
2.63
9.81
7.92
25.11 21.39
85.41 87.06
1,028 1,212

7.77
11.41
0.00
0.39
1.20
3.35
9.03
20.97
87.89
1,383

9.35
13.01
0.00
0.53
1.62
4.49
11.28
24.21
92.36
1,492

10.57
12.26
0.00
0.77
2.28
6.31
13.94
25.98
94.65
1,647

8.42
9.90
0.00
0.43
1.60
4.97
11.74
21.48
92.44
1,753

11.24
11.10
0.00
0.93
2.72
7.69
16.46
26.97
79.96
1,966

(continued)

⎧1
⎪
⎪1
(7c) It = ⎨
⎪0
⎪0
⎩

with probability p11 if It −1 = 1
with probability p21 if It−1 = 0
with probability p12 if It−1 = 1

with probability p22 if It−1 = 0

This specification is similar to the well-known “mixture of distributions” model.
However, unlike standard mixture models, the regime-switching model is not independently distributed over time unless p11 = p21. Once estimated, forecasts of changes
in regime can be readily obtained as well as forecasts of Rt itself. In particular,
70

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Table 7 (continued)
Liquidation Probabilities of Logit Models for
Annual Hedge Fund Liquidations, January 1994–August 2004
1994

1995

1996

1997

Model 5
1998
1999

2000

2001

2002

2003

2004

Mean
SD
Min
10%
25%
50%
75%
90%
Max
Count

1.06
3.28
0.00
0.00
0.02
0.07
0.52
2.61
35.62
357

2.22
6.01
0.00
0.01
0.04
0.16
1.25
5.85
42.56
483

4.30
10.97
0.00
0.02
0.09
0.36
2.61
11.24
76.54
629

3.43
8.70
0.00
0.02
0.10
0.45
2.26
9.12
86.91
773

Live funds
4.70
4.05
9.51
8.87
0.00
0.00
0.06
0.04
0.27
0.23
1.03
0.96
4.03
3.22
14.21 10.09
77.72 80.45
924 1,083

3.80
7.72
0.00
0.07
0.33
1.18
3.49
9.88
75.95
1,207

3.40
6.76
0.00
0.07
0.33
1.26
3.63
8.10
91.82
1,317

4.07
6.58
0.00
0.09
0.44
1.74
4.75
10.52
73.06
1,480

4.45
6.33
0.00
0.07
0.43
2.04
6.01
12.03
81.10
1,595

1.76
2.70
0.00
0.03
0.15
0.72
2.31
4.71
29.28
1,898

Mean
SD
Min
10%
25%
50%
75%
90%
Max
Count

24.23
24.12
1.00
5.31
11.79
18.02
26.24
48.95
64.10
5

23.50
20.12
4.92
5.53
7.99
17.66
32.58
51.10
69.64
14

34.07
25.19
1.88
5.25
11.28
33.94
54.36
68.87
82.29
41

42.30
26.95
1.49
8.61
21.29
37.54
64.53
80.97
93.17
46

Graveyard funds
36.17 31.46
25.12 21.96
0.00
0.11
4.49
2.12
15.56 12.66
28.92 30.16
60.14 46.31
69.54 64.68
87.67 89.00
68
64

32.55
22.47
0.02
3.95
15.91
27.57
48.38
61.91
90.90
68

22.82
19.84
0.51
2.00
6.43
19.11
33.10
55.75
76.34
58

20.68
18.94
0.03
2.61
5.29
14.32
33.19
46.84
90.02
76

20.18
16.27
0.03
3.02
6.42
14.03
30.61
43.06
67.86
89

4.60
6.20
0.04
0.13
0.97
3.16
5.51
10.17
33.31
35

Mean
SD
Min
10%
25%
50%
75%
90%
Max
Count

1.38
4.94
0.00
0.00
0.02
0.08
0.56
3.06
64.10
362

2.82
7.62
0.00
0.01
0.04
0.19
1.38
7.02
69.64
497

6.12
14.21
0.00
0.02
0.10
0.43
3.58
19.05
82.29
670

5.62
13.84
0.00
0.03
0.11
0.54
3.02
16.84
93.17
819

Combined funds
6.85
5.58
13.79 11.85
0.00
0.00
0.06
0.05
0.30
0.24
1.24
1.06
5.57
4.27
22.27 17.07
87.67 89.00
992 1,147

5.33
11.17
0.00
0.07
0.35
1.32
4.40
15.37
90.90
1,275

4.22
8.68
0.00
0.07
0.35
1.42
4.15
9.65
91.82
1,375

4.88
8.44
0.00
0.09
0.48
1.93
5.36
12.50
90.02
1,556

5.29
8.01
0.00
0.08
0.49
2.28
6.63
13.79
81.10
1,684

1.81
2.82
0.00
0.03
0.15
0.73
2.36
4.85
33.31
1,933

Note: The summary statistics use annual observations of the liquidation status of individual hedge funds in the TASS database.

because the k-step transition matrix of a Markov chain is simply given by P k, the conditional probability of the regime It+k given date-t data Rt ≡ (Rt, Rt–1,…, R1) takes on a
particularly simple form:
(8a) Prob(It+k = 1|Rt) = π1 + (p11 – p21)k [Prob(It = 1|Rt ) – π1],
(8b)

π1 ≡

p21
,
p12 + p21

where Prob(It = 1|Rt) is the probability that the date-t regime is 1 given the historical
data up to and including date t (this is a by-product of the maximum-likelihood

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Table 8
Maximum Likelihood Estimates of a Two-State Regime-Switching Model for
CSFB/Tremont Hedge Fund Indexes, January 1994–August 2004

Index

p11 (%)

p21 (%)

p12 (%)

p22 (%)

Hedge funds
100.0
Convertible arbitrage
89.9
Dedicated short-seller
23.5
Emerging markets
100.0
Equity market neutral
95.0
Event driven
98.0
Distressed
97.9
Event-driven multistrategy
98.7
Risk arbitrage
89.4
Fixed income arbitrage
95.6
Global macro
100.0
Long/short equity
98.5
Managed futures
32.0
Multistrategy
98.2

1.2
17.9
12.6
1.2
2.4
45.0
58.0
38.4
25.6
29.8
1.2
2.5
22.2
25.0

0.0
10.1
76.5
0.0
5.0
2.0
2.1
1.3
10.6
4.4
0.0
1.5
68.0
1.8

98.8
82.1
87.4
98.8
97.6
55.0
42.0
61.6
74.4
70.2
98.8
97.5
77.8
75.0

Annualized mean
State 1 State 2
(%)
(%)
6.8
16.1
–76.2
11.5
4.4
13.3
15.2
12.0
9.6
10.0
13.6
6.1
–6.0
10.8

12.4
–1.6
11.7
6.6
13.8
–47.0
–57.5
–55.2
3.1
–12.2
14.0
21.1
10.7
–7.6

Annualized SD
State 1 State 2
(%)
(%)
2.9
1.9 6
2.3
8.2
2.1
3.8
4.8
4.5
2.7
1.9
3.2
6.3
3.8
3.2

9.9
.1
16.5
20.3
3.1
14.0
15.6
15.0
6.9
6.6
14.2
15.3
13.7
9.2

Log(L)
323.6
404.0
208.5
218.0
435.1
377.0
349.4
363.6
391.8
442.3
286.3
285.0
252.1
387.9

Note: Highlighted rows indicate unreliable maximum likelihood estimates (either nonconvergence or multiple local maxima).

estimation procedure). Using similar recursions of the Markov chain, the conditional
expectation of Rt+k can be readily derived as
(9a) E[Rt+k|Rt] = a′tPkµ;
(9b)

at = [Prob(It = 1|Rt) Prob(It = 2|Rt)]′;

(9c)

µ ≡ [µ1 µ2]′.

Table 8 reports the maximum-likelihood estimates of the means and standard
deviations in each of two states for the fourteen CSFB/Tremont hedge fund indexes,
as well as the transition probabilities for the two states. Note that two rows in Table 8
are shaded—dedicated short-seller and managed futures—because the maximumlikelihood estimation procedure did not converge properly for these two categories,
implying that the regime-switching process may not be a good model of their returns.
The remaining twelve series yielded well-defined parameter estimates, and, by convention, we denote by state 1 the lower-volatility state.
Consider the second row, corresponding to the convertible arbitrage index. The
parameter estimates indicate that in state 1 this index has an expected return of 16.1
percent with a volatility of 1.9 percent, but in state 2 the expected return is –1.6 percent with a volatility of 6.1 percent. The latter state is clearly a crisis state for convertible arbitrage, while the former is a more normal state. The other hedge fund
indexes have similar parameter estimates—the low-volatility state is typically paired
with higher means, and the high-volatility state is paired with lower means. While
such pairings may seem natural for hedge funds, there are three exceptions to this
rule: For equity market neutral, global macro, and long/short equity, the higher

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Figure 3
Monthly Returns and Probabilities of the High-Volatility State for the
CSFB/Tremont Fixed-Income Arbitrage Hedge Fund Index, January 1994–August 2004
3

100

2
90

80
70

–2
–3

–4
–5

Probability of high
volatility (percent)

–6
–7

100

Probability (percent)

Return (percent)

0
–1

50
10

0
1994

1996

1998

2000

2002

2004

volatility state has higher expected returns. This outcome suggests that for these
strategies, volatility may be a necessary ingredient for their expected returns.
From these parameter estimates, it is possible to estimate the probability of being
in state 1 or 2 at each point in time for each hedge fund index. For example, in Figure 3
we plot the estimated probabilities of being in state 2, the high-volatility state, for the
fixed-income arbitrage index for each month from January 1994 to August 2004. We
see that this probability begins to increase in the months leading up to August 1998 and
hits 100 percent in August and several months thereafter. However, this is not an isolated event but occurs on several occasions both before and after August 1998.
To develop an aggregate measure of systemic risk based on this regime-switching
model, we propose summing the state-2 probabilities across all hedge fund indexes
every month to yield a time series that captures the likelihood of being in low-mean
periods. Of course, the summed probabilities—even if renormalized to lie in the
unit interval—cannot be interpreted formally as a probability because the regimeswitching process was specified individually for each index, not jointly across all
indexes. Therefore, the interpretation of the low-mean state for convertible arbitrage may be quite different than the interpretation of the low-mean state for equity
market neutral. Nevertheless, as an aggregate measure of the state of the hedge
fund industry, the summed probabilities may contain useful information about systemic risk exposures. Figure 4 contains this indicator. The low-mean indicator has
local maxima in 1994 and 1998 as expected, but there is a stronger peak around
2002, largely due to equity market neutral, global macro, and long/short equity.
This pattern corresponds remarkably well to the common wisdom that, over the
past two years, these three strategy classes have underperformed for a variety of

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Figure 4
Regime-Switching Probabilities of Low-Mean States for
CSFB/Tremont Hedge Fund Indexes, January 1994–August 2004
10
Convertible arbitrage
Event driven
Risk arbitrage
Long/short equity

Emerging markets
Distressed
Fixed-income arbitrage
Multistrategy

Equity market neutral
Event-driven multistrategy
Global macro

Correlation (percent)

0
1994

1996

1998

2000

2002

2004

Note: The figure shows the sum of monthly regime-switching model estimates of the probability of being in the low-mean state for eleven
CSFB/Tremont hedge fund indexes.

reasons.20 The implications of Figure 4 for systemic risk are clear: the probabilities
of being in low-mean regimes have increased for a number of hedge fund indexes,
which may foreshadow increased leverage for funds in these categories as well as
fund outflows in the coming months, both of which would place additional stress on
the industry, leading to an increase in systemic risk.

The Current Outlook
A definitive assessment of the systemic risks posed by hedge funds requires certain
data that are currently unavailable and are unlikely to become available in the near
future—for example, counterparty credit exposures, the net degree of leverage of
hedge fund managers and investors, the gross amount of structured products involving hedge funds, etc. Therefore, we cannot determine the magnitude of current systemic risk exposures with any degree of accuracy. However, based on the analytics
developed in this study, there are a few tentative inferences that we can draw.
1. The hedge fund industry has grown tremendously over the last few years, fueled
by the demand for higher returns in the face of stock-market declines and mounting pension-fund liabilities. These massive fund inflows have had a material impact
on hedge fund returns and risks in recent years, as evidenced by changes in correlations, reduced performance, and increased illiquidity as measured by the
weighted autocorrelation ρ*t .
2. Mean and median liquidation probabilities for hedge funds have increased in
2004, based on logit estimates that link several factors to the liquidation probability of a given hedge fund, including past performance, AUM, fund flows, and
age. In particular, our estimates imply that the average liquidation probability for
funds in 2004 is over 11 percent—higher than the historical unconditional attri-

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

tion rate of 8.8 percent. A higher attrition rate is not surprising for a rapidly growing industry, but it may foreshadow potential instabilities that can be triggered by
seemingly innocuous market events.
3. The banking sector is exposed to hedge fund risks, especially smaller institutions,
but the largest banks are also exposed through proprietary trading activities,
credit arrangements and structured products, and prime brokerage services.
4. The risks facing hedge funds are nonlinear and more complex than those facing
traditional asset classes. Because of the dynamic nature of hedge fund investment
strategies and the impact of fund flows on leverage and performance, hedge fund
risk models require more sophisticated analytics and more sophisticated users.
5. The sum of our regime-switching models’ low-mean state probabilities is one proxy
for the aggregate level of distress in the hedge fund sector. Recent measurements
suggest that we may be entering a challenging period of lower expected returns.
This new regime, coupled with the recent uptrend in the weighted autocorrelation ρ*t and the increased mean and median liquidation probabilities for hedge
funds in 2004 from our logit model, implies that systemic risk is increasing.
We hasten to qualify our tentative conclusions by emphasizing the speculative nature
of these inferences, and we hope that our analysis spurs additional research and data
collection to refine both the analytics and the empirical measurement of systemic
risk in the hedge fund industry. As with all risk management challenges, we should
hope for the best and prepare for the worst. The question is, How?
One possibility, put forward by Getmansky, Lo, and Mei (2004), is to create an
independent organization along the lines of the National Transportation Safety Board
(NTSB) to sift through the wreckage of all major hedge fund collapses, ultimately
producing a publicly available report that documents the specific causes of the collapse, along with recommendations on how to avoid similar disasters in the future.
Although there may be common themes in the demise of many hedge funds—too
much leverage, too concentrated a portfolio, operational failures, securities fraud, or
insufficient AUM—each liquidation has its own unique circumstances and is an
opportunity for hedge fund managers and investors to learn and improve.
In the event of an airplane crash, the NTSB assembles a team of engineers and
flight-safety experts who are immediately dispatched to the crash site to conduct a
thorough investigation, including interviewing witnesses, poring over historical flight
logs and maintenance records, and sifting through the wreckage to recover the flight
recorder or “black box” and, if necessary, reassembling the aircraft from its parts to
determine the ultimate cause of the crash. Once its work is completed, the NTSB
publishes a report summarizing the team’s investigation, concluding with specific
recommendations for avoiding future occurrences of this type of accident. The report
is entered into a searchable database that is available to the general public (see
www.ntsb.gov/ntsb/query.asp), and this kind of information has been one of the major
factors underlying the remarkable safety record of commercial air travel.
For example, it is now current practice to spray airplanes with deicing fluid just
prior to takeoff when the temperature is near freezing and it is raining or snowing.
This procedure was instituted in the aftermath of USAir Flight 405’s crash on March 22,
1992. Flight 405 stalled just after becoming airborne because of ice on its wings,
20. Large fund flows into these strategies and changes in equity markets such as decimalization, the
rise of electronic communication networks (ECNs), automated trading, and Regulation FD are
often cited as reasons for the decreased profitability of these strategies.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

despite the fact that deicing fluid was applied before it left its gate. Apparently, Flight
405’s takeoff was delayed because of air traffic, and ice reaccumulated on its wings
while it waited for a departure slot on the runway in the freezing rain. The NTSB
Aircraft Accident Report AAR-93/02—published February 17, 1993, and available
through several Internet sites—contains a sobering summary of the NTSB’s findings:
The National Transportation Safety Board determines that the probable causes
of this accident were the failure of the airline industry and the Federal Aviation
Administration to provide flightcrews with procedures, requirements, and criteria
compatible with departure delays in conditions conducive to airframe icing and
the decision by the flightcrew to take off without positive assurance that the airplane’s wings were free of ice accumulation after 35 minutes of exposure to precipitation following deicing. The ice contamination on the wings resulted in an
aerodynamic stall and loss of control after liftoff. Contributing to the cause of the
accident were the inappropriate procedures used by, and inadequate coordination
between, the flightcrew that led to a takeoff rotation at a lower than prescribed
air speed. (Report AAR-93/02, page vi)

Current deicing procedures have no doubt saved many lives thanks to NTSB Report
AAR-93/02, but this particular innovation was paid for by the lives of the twenty-seven
individuals who did not survive the crash of Flight 405. Imagine the waste if the NTSB
did not investigate this tragedy and produce concrete recommendations to prevent
such situations from happening again.
Hedge fund liquidations are, of course, considerably less dire, generally involving
no loss of life. However, as more pension funds make allocations to hedge funds, and
as the “retailization” of hedge funds continues, losses in the hedge fund industry may
have more significant implications for individual investors, in some cases threatening
retirement wealth and basic living standards. Moreover, the spillover effects of an
industrywide shock to hedge funds should not be underestimated, as the events surrounding LTCM in the fall of 1998 illustrated. For these reasons, a “Capital Markets
Safety Board” (CMSB) dedicated to investigating, reporting, and archiving the “accidents” of the hedge fund industry—and the financial services sector more generally—
may yield significant social benefits in much the same way that the NTSB has improved
transportation safety enormously for all air travelers. By maintaining teams of experienced professionals—forensic accountants, financial engineers from industry and
academia, and securities and tax attorneys—who work together on a regular basis to
investigate a number of hedge fund liquidations, this investigative body would be able
to determine quickly and accurately how each liquidation came about, and the resulting reports would be an invaluable source of ideas for improving financial markets
and avoiding future liquidations of a similar nature.
Of course, formal government investigations of major financial events do occur
from time to time, as in the April 1999 report of the President’s Working Group on
Financial Markets. However, this interagency report was put together on an ad hoc
basis with committee members that had not worked together previously and regularly on forensic investigations of this kind. With multiple agencies involved, and none
in charge of the investigation, the administrative overhead becomes more significant.
Although any thorough investigation of the financial services sector is likely to
involve the SEC, the Commodity Futures Trading Commission, the U.S. Treasury, and
the Federal Reserve—and interagency cooperation should be promoted—there are
important operational advantages in tasking a single independent office with the

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

responsibility for coordinating all such investigations and serving as a repository for
the expertise in conducting forensic examinations of financial incidents.
The establishment of the CMSB will not be inexpensive. Currently, regulatory
agencies like the SEC are understaffed and overburdened, and this condition is likely
to worsen as financial markets grow in size
Valuable lessons could be garnered from a
and complexity. In addition, the lure of the
private sector makes it challenging for
systematic analysis of financial incidents
government agencies to attract and retain
and the public dissemination of recommenindividuals with expertise in these highly
dations by seasoned professionals that
employable fields. Individuals trained in
forensic accounting, financial engineering,
review multiple cases each year.
and securities law now command substantial premiums on Wall Street over government pay scales. Although the typical publicsector employee is likely to be motivated more by civic duty than financial gain, it
would be unrealistic to build an organization on altruism alone.
However, the cost of an independent CMSB is more than justified by the valuable
lessons that would be garnered from a systematic analysis of financial incidents and
the public dissemination of recommendations by seasoned professionals that review
multiple cases each year. The benefits would accrue not only to the wealthy—which
is currently how the hedge fund industry is tilted—but would also flow to retail
investors in the form of more stable financial markets, greater liquidity, reduced borrowing and lending costs as a result of decreased systemic risk exposures, and a
wider variety of investment choices available to a larger segment of the population
because of increased transparency, oversight, and ultimately, financial security. It is
unrealistic to expect that market crashes, panics, collapses, and fraud will ever be
completely eliminated from our capital markets, but we should avoid compounding
our mistakes by failing to learn from them.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

Appendix
A Description of TASS Hedge Fund Categories

he following is a list of category descriptions, taken directly from TASS documentation, that define the criteria used by TASS in
assigning funds in their database to one of
eleven possible categories:

Convertible arbitrage. This strategy is identified by hedge investing in the convertible
securities of a company. A typical investment is
to be long the convertible bond and short the
common stock of the same company. Positions
are designed to generate profits from the fixedincome security as well as the short sale of
stock while protecting principal from market
moves.
Dedicated short-seller. Dedicated shortsellers were once a robust category of hedge
funds before the long bull market rendered the
strategy difficult to implement. A new category,
short biased, has emerged. The strategy is to
maintain net short as opposed to pure short
exposure. Short biased managers take short
positions in mostly equities and derivatives.
The short bias of a manager’s portfolio must be
constantly greater than zero to be classified in
this category.
Emerging markets. This strategy involves
equity or fixed-income investing in emerging
markets around the world. Because many
emerging markets do not allow short selling or
offer viable futures or other derivative products
with which to hedge, emerging market investing often employs a long-only strategy.
Equity market neutral. This investment
strategy is designed to exploit equity market
inefficiencies and usually involves being simultaneously long and short matched equity portfolios of the same size within a country. Market
neutral portfolios are designed to be either beta
or currency neutral or both. Well-designed
portfolios typically control for industry, sector,
market capitalization, and other exposures.
Leverage is often applied to enhance returns.

ECONOMIC REVIEW

Fourth Quarter 2006

Event driven. This strategy is defined as “special situations” investing designed to capture
price movement generated by a significant
pending corporate event such as a merger, corporate restructuring, liquidation, bankruptcy,
or reorganization. There are three popular subcategories in event-driven strategies: risk
(merger) arbitrage, distressed/high-yield securities, and Regulation D.
Fixed-income arbitrage. The fixed-income
arbitrageur aims to profit from price anomalies
between related interest rate securities. Most
managers trade globally with a goal of generating steady returns with low volatility. This category includes interest rate swap arbitrage, U.S.
and non-U.S. government bond arbitrage, forward yield curve arbitrage, and mortgage-backedsecurities arbitrage. The mortgage-backed market
is primarily U.S.-based, over-the-counter, and
particularly complex.
Global macro. Global macro managers carry
long and short positions in any of the world’s
major capital or derivative markets. These positions reflect their views on overall market direction as influenced by major economic trends
and/or events. The portfolios of these funds can
include stocks, bonds, currencies, and commodities in the form of cash or derivative instruments.
Most funds invest globally in both developed
and emerging markets.
Long/short equity. This directional strategy
involves equity-oriented investing on both the
long and short sides of the market. The objective is not to be market neutral. Managers have
the ability to shift from value to growth, from
small to medium to large capitalization stocks,
and from a net long position to a net short position. Managers may use futures and options to
hedge. The focus may be regional, such as
long/short U.S. or European equity, or sector
specific, such as long and short technology or
health care stocks. Long/short equity funds
tend to build and hold portfolios that are sub-

F E D E R A L R E S E R V E B A N K O F AT L A N TA

stantially more concentrated than those of traditional stock funds.
Managed futures. This strategy invests in listed
financial and commodity futures markets and currency markets around the world. The managers are
usually referred to as commodity trading advisers,
or CTAs. Trading disciplines are generally systematic or discretionary. Systematic traders tend to use
price and market-specific information (often technical) to make trading decisions, while discretionary managers use a judgmental approach.
Multistrategy. The funds in this category are
characterized by their ability to dynamically

allocate capital among strategies falling within
several traditional hedge fund disciplines. The
use of many strategies, and the ability to reallocate capital between them in response to market opportunities, means that such funds are
not easily assigned to any traditional category.
The multistrategy category also includes funds
employing unique strategies that do not fall
under any of the other descriptions.
Fund of funds. A “multi manager” fund will
employ the services of two or more trading
advisers or hedge funds who will be allocated
cash by the trading manager to trade on behalf
of the fund.

REFERENCES
Agarwal, Vikas, Naveen D. Daniel, and Narayan Naik.
2004. Flows, performance and managerial incentives in
hedge funds. EFA 2003 Annual Conference Paper No.
501 (July 22). Available at <http://ssrn.com/abstract=
424369> (November 22, 2006).
Ang, Andrew, and Geert Bekaert. 2004. How regimes
affect asset allocation. Financial Analysts Journal
60, no. 2:86–99.

Campbell, John Y., Andrew W. Lo, and A. Craig MacKinlay.
1996. The econometrics of financial markets. Princeton,
N.J.: Princeton University Press.
Carpenter, Jennifer N., and Anthony W. Lynch. 1999.
Survivorship bias and attrition effects in measures of
performance persistence. Journal of Financial
Economics 54, no. 3:337–74.

Baquero, Guillermo, Jenke ter Horst, and Marno Verbeek.
2005. Survival, look-ahead bias and persistence in hedge
fund performance. Journal of Financial and Quantitative Analysis 40, no. 3:493–517.

Chan, Nicholas, Mila Getmansky, Shane M. Haas, and
Andrew Lo. Forthcoming. Systemic risk and hedge
funds. In The risks of financial institutions and the
financial sector, edited by Mark Carey and René M.
Stulz. Chicago: University of Chicago Press.

Bookstaber, Richard. 1999. A framework for understanding market crisis. In Risk management: Principles and practices. Charlottesville, Va.: Association
for Investment Management and Research.

Chandar, Nandini, and Robert Bricker. 2002. Incentives,
discretion, and asset valuation in closed-end mutual
funds. Journal of Accounting Research 40, no. 4:
1037–70.

———. 2000. Understanding and monitoring the liquidity
crisis cycle. Financial Analysts Journal 56, no. 5:17–22.

Chevalier, Judith, and Glenn Ellison. 1997. Risk taking
by mutual funds as a response to incentives. Journal
of Political Economy 105, no. 6:1167–1200.

Brown, Stephen J., William N. Goetzmann, and Robert
G. Ibbotson. 1999. Offshore hedge funds: Survival and
performance, 1989–95. Journal of Business 72, no.
1:91–118.

Cox, David R., and E. Joyce Snell, eds. 1989. The
analysis of binary data, 2d ed. London: Chapman
and Hall.

Brown, Stephen J., William N. Goetzmann, Robert G.
Ibbotson, and Stephen A. Ross. 1992. Survivorship bias
in performance studies. Review of Financial Studies
5, no. 4:553–80.

Farmer, J. Doyne, and Andrew W. Lo. 1999. Frontiers
of finance: Evolution and efficient markets. Proceedings of the National Academy of Sciences 96,
9991–92.

Brown, Stephen J., William N. Goetzmann, and James
M. Park. 2001. Careers and survival: Competition and
risk in the hedge fund and CTA industry. Journal of
Finance 56, no. 5:1869–86.

Fung, William, and David Hsieh. 1997. Survivorship
bias and investment style in the returns of CTAs.
Journal of Portfolio Management 24, no. 1:30–41.

ECONOMIC REVIEW

Fourth Quarter 2006

F E D E R A L R E S E R V E B A N K O F AT L A N TA

———. 2000. Performance characteristics of hedge
funds and commodity funds: Natural vs. spurious biases.
Journal of Financial and Quantitative Analysis 35,
no. 3:291–307.
Getmansky, Mila, Andrew Lo, and Igor Makarov. 2004.
An econometric model of serial correlation and illiquidity in hedge fund returns. Journal of Financial
Economics 74, no. 3:529–609.
Getmansky, Mila, Andrew Lo, and Shauna X. Mei.
2004. Sifting through the wreckage: Lessons from
recent hedge-fund liquidations. Journal of Investment Management 2, no. 4:6–38.
Hamilton, James D. 1989. A new approach to the economic analysis of nonstationary time series and the
business cycle. Econometrica 57, no. 2:357–84.
———. 1990. Analysis of time series subject to changes
in regime. Journal of Econometrics 45, nos. 1–2:39–70.
Hendricks, Darryll, Jayendu Patel, and Richard
Zeckhauser. 1997. The J-shape of performance persistence given survivorship bias. Review of Economics
and Statistics 79, no. 2:161–66.
Horst, Jenke ter, Theo E. Nijman, and Marno Verbeek.
2001. Eliminating look-ahead bias in evaluating persistence in mutual fund performance. Journal of Empirical Finance 8, no. 4:345–73.
Kadlec, Gregory B., and Douglas M. Patterson. 1999. A
transactions data analysis of nonsynchronous trading.
Review of Financial Studies 12, no. 3:609–30.
Kao, Duen-Li. 2000. Estimating and pricing credit risk:
An overview. Financial Analysts Journal 56, no. 4:
50–66.
Kendall, Maurice G., Alan Stuart, and J. Keith Ord.
1983. The advanced theory of statistics, vol. 3, 4th ed.
New York: Macmillan.
Leamer, Edward E. 1978. Specification searches.
New York: John Wiley and Sons.
Liang, Bing. 2003. The accuracy of hedge fund returns.
Journal of Portfolio Management 29, no. 3:111–22.
Ljung, G.M., and G.E.P. Box. 1978. On a measure of
lack of fit in time series models. Biometrika 65, no. 2:
297–303.
Lo, Andrew W. 1999. The three p’s of total risk management. Financial Analysts Journal 55, no. 1:
13–27.

ECONOMIC REVIEW

Fourth Quarter 2006

———. 2001. Risk management for hedge funds:
Introduction and overview. Financial Analysts
Journal 57, no. 6:16–33.
———. 2002. The statistics of Sharpe ratios. Financial
Analysts Journal 58, no. 4:36–52.
———. 2004. The adaptive markets hypothesis: Market
efficiency from an evolutionary perspective. Journal
of Portfolio Management 30 (30th anniversary issue):
15–29.
Lo, Andrew W., and A. Craig MacKinlay. 1988. Stock
market prices do not follow random walks: Evidence
from a simple specification test. Review of Financial
Studies 1, no. 1:41–66.
———. 1990. An econometric analysis of nonsynchronous trading. Journal of Econometrics 45,
nos. 1–2:181–211.
Maddala, G.S. 1983. Limited-dependent and qualitative variables in econometrics. Cambridge: Cambridge
University Press.
Mishkin, Frederic S. 1997. The economics of money,
banking, and financial markets, 5th ed. Reading,
Mass.: Addison-Wesley.
Nagelkerke, N.J.D. 1991. A note on a general definition
of the coefficient of determination. Biometrika 78, no. 3:
691–92.
President’s Working Group on Financial Markets. 1999.
Hedge Funds, Leverage, and the Lessons of LongTerm Capital Management: Report of the President’s
Working Group on Financial Markets. April.
Samuelson, Paul A. 1965. Proof that properly anticipated prices fluctuate randomly. Industrial Management Review 6 (Spring): 41–49.
Schneeweis, Thomas, Richard Spurgin, and David
McCarthy. 1996. Survivor bias in commodity trading
advisor performance. Journal of Futures Markets 16,
no. 7:757–72.
Sirri, Erik R., and Peter Tufano. 1998. Costly search
and mutual fund flows. Journal of Finance 53, no. 5:
1589–1622.
Watts, Duncan J. 1999. Small worlds: The dynamics
of networks between order and randomness.
Princeton, N.J.: Princeton University Press.
Watts, Duncan J., and Steven H. Strogatz. 1998. Collective dynamics of “small-world” networks. Nature 393,
no. 6684:440–42.

Full text of Economic Review (Federal Reserve Bank of Atlanta) : Fourth Quarter 2006, Volume 91, Number 4

FRASER