ArticlePDF Available

Challenges arising from alternative investment management.



Alternative investment management differs from traditional asset management in a number of respects. First, it is distinct in terms of both its targets – aiming to achieve an absolute performance, regardless of trends in underlying markets – and its strategies, in particular exploiting inefficiencies in the valuation of financial assets via opportunistic and discretionary positions. It also differs in terms of the financial techniques implemented, e.g. the extensive use made of leverage, derivatives and short selling, and the specific investment vehicles used (ad hoc structures such as hedge funds that are not bound by ordinary law in the way traditional investment vehicles are). These particularities, alongside the fact that the alternative investment universe is somewhat opaque, make it difficult to measure a fund’s risks or a fund manager’s performance. Specific measurement tools are therefore required, which differ from those commonly used in traditional asset management. Over the past few years, the alternative investment management, a diverse and rapidly-evolving universe, has enjoyed a spectacular development, which is illustrated by the sharp rise in the amounts under management and the proliferation of investment vehicles offered to an increasingly broad investor base. In view of the specific nature of alternative fund managers’ modus operandi, the flourishing of the alternative investment industry raises questions as to its implications in terms of financial stability. It also raises new issues regarding the division of roles between market participants and supervisory authorities in the organisation and monitoring of this asset management sector.
Banque de FranceFinancial Stability Review No. 3 November 2003 1
Challenges arising from alternative
investment management
François HAAS
Banque de France
Market and Financial Stability Research
Professor of Finance, EDHEC
Misys Asset Management Systems
Research Engineer, EDHEC
Over the past few years the alternative
investment industry has developed
spectacularly, attracting so much investor
interest that it appears at times to have taken on the
proportions of a “fad”. This article sets out to chart the
development of this phenomenon. We will start out by
attempting to define the broad outlines and trends of
this rapidly evolving industry. We will then try to show
why this style of management requires specific risk
analysis techniques and performance indicators. This
will enable us to assess the impact of alternative
investment management on financial market dynamics,
and identify the conditions under which these
management strategies might “usefully” contribute to
the functioning of financial markets and become a lasting
feature in the world of asset management.
Alternative investment management differs from traditional asset management in a number of
respects. First, it is distinct in terms of both its targets – aiming to achieve an absolute performance,
regardless of trends in underlying markets – and its strategies, in particular exploiting inefficiencies
in the valuation of financial assets via opportunistic and discretionary positions. It also differs in
terms of the financial techniques implemented, e.g. the extensive use made of leverage, derivatives
and short selling, and the specific investment vehicles used (ad hoc structures such as hedge funds
that are not bound by ordinary law in the way traditional investment vehicles are). These
particularities, alongside the fact that the alternative investment universe is somewhat opaque,
make it difficult to measure a fund’s risks or a fund manager’s performance. Specific measurement
tools are therefore required, which differ from those commonly used in traditional asset management.
Over the past few years, the alternative investment management, a diverse and rapidly-evolving
universe, has enjoyed a spectacular development, which is illustrated by the sharp rise in the amounts
under management and the proliferation of investment vehicles offered to an increasingly broad
investor base. In view of the specific nature of alternative fund managers’ modus operandi, the
flourishing of the alternative investment industry raises questions as to its implications in terms of
financial stability. It also raises new issues regarding the division of roles between market participants
and supervisory authorities in the organisation and monitoring of this asset management sector.
Challenges arising from alternative investment management
2 Banque de FranceFinancial Stability Review No. 3 November 2003
1| The universe of alternative investment management
Given the diversity and evolving nature of
this industry, it is not easy to define what is
commonly understood by “alternative investment
management”. Before analysing the trends currently
characterising this universe, we will attempt to
determine its contours.
1|1 An attempted definition
Alternative investment management can be defined
in terms of how it differs from “traditional” asset
management: the targets set, the instruments and
techniques used, and the investment strategies
Investment targets
In “traditional” asset management, as practised
by mutual funds and institutional fund managers,
the investment rationale is typically a
relative performance approach: investments under
management in this form are essentially, if not
exclusively, allocated to pre-specified financial asset
classes, at a medium- to long-term horizon.
Consequently, the performance of a portfolio
managed in this way, and the risk to which it is
exposed, will depend primarily on the trends of the
underlying markets and, secondly, on the fund
manager’s ability to allocate assets appropriately.
This performance will be measured against an
overall or sectoral benchmark index, considered to
be representative of the markets in which the assets
are invested (money markets, bond markets, equity
markets, domestic or international markets). The
fund manager aims to achieve a performance that is
either as close as possible to that of the benchmark
index (index-linked or passive management) or to
beat this index (active management).
Conversely, in alternative investment management,
an absolute positive performance target is set,
i.e. a performance uncorrelated with that of the
underlying asset classes. In principle, managers set
out to achieve a performance that is (at least
partially) independent of the intrinsic performance
of financial markets: investors forgo structural
returns associated with “long-only” investment
positions (the risk premium adjusted for the level of
market exposure – beta 1), in exchange for protection
against adverse movements in markets (directional
risk). Investors are thus directly exposed to the
quality of the fund manager 2.
While the traditional asset management approach is
based on efficient market and optimal market
portfolio hypotheses, the alternative investment
approach looks for inefficiencies in financial markets
with a view to exploiting them: the alternative
investment approach aims to capture alpha rather
than beta, i.e. to outperform the market.
Techniques and instruments
Alternative investment management, like traditional
asset management, uses conventional financial
instruments such as money market paper, bonds and
equities. However, it also uses less conventional, and
generally less liquid, categories of assets (unlisted
securities, commodities, or even real estate assets).
Furthermore it employs, in the usual manner, and
sometimes in significant proportions, a complete
range of outright and conditional derivatives, as well
as specific financial techniques such as short selling.
While leverage 3 is not specific to alternative
investment, nor systematically used 4, it is part of
alternative fund managers’ array of everyday
1Broadly speaking, beta is the measure of the volatility of an individual (portfolio) security compared with that of the market as a whole, i.e. its
systematic risk. Alpha represents the excess returns on an individual (portfolio) security, which is not explained by the model. For a more
detailed definition, see Box 1 below.
2Hence, the performance structure of hedge fund managers is directly linked to their performance: on average, on the basis of a number of market
sources, we can estimate that fund managers receive almost 18% of profits generated, as well as management fees of 1% to 1.5% of the amount
of assets under management.
3In this paper, we define leverage as the use of borrowed resources or the use of the above-mentioned financial techniques and instruments with
a view to increasing the size of the positions taken.
4In particular when exploiting minor arbitrage opportunities.
Challenges arising from alternative investment management
Banque de FranceFinancial Stability Review No. 3 November 2003 3
Alternative funds implement a wide range of
investment strategies. While these strategies are
flexible, and many different combinations exist, we
can basically identify four major investment styles,
of which there are numerous variations.
Long/short strategies involve taking simultaneously
short and long positions in different securities,
some considered undervalued and others
overvalued on the basis of a fundamental or
technical analysis, and then exploiting this
market anomaly. The overall position can be
directional – skewed in favour of one or other of
the positions, or non-directional (market neutral).
In the latter case, the profit generated is not
dependent on the performance of the market as
a whole (both securities may appreciate or
depreciate simultaneously) but solely on the
relative performance of each of the positions.
Arbitrage 5 and relative value strategies are focused
on detecting and exploiting perceived anomalies
in the relative pricing of financial assets or in
the statistical relationships linking different
assets. On the basis of this principle, arbitrage
strategies between different segments of the yield
curve or bond markets exploit yield differentials
and, in particular, the mean return tendency of
these spreads. Also of note are convertible
securities arbitrages, which aim to exploit
potential pricing differences between the security
itself and its different components 6 or take
position in the underlying factors that determine
the value of these particular securities. Another
technique employed is “capital structure
arbitrage”, whose objective is to exploit any
differences in the pricing of an issuer’s liabilities
(between its debts according to their ranking, or
between debt and equity securities). As different
as they may be, these strategies share one
common feature: the intensive use of quantitative
techniques and sophisticated mathematical
modelling to identify arbitrage opportunities. The
latter are systematically exploited by the taking
of both long positions and short selling. Given
that arbitrage opportunities are often minor, the
positions taken by these funds are frequently
highly leveraged.
Event-driven and special situation strategies can be
defined as positions taken on the developments
of a situation of a particular company, or on the
probability of the occurrence of a particular event
in the life of a company. Some funds, sometimes
known as “vulture funds”, specialise in the debt
of restructuring companies (distressed debt),
which is often undervalued due to the fact that
many investors find it impossible to hold this
type of paper in their portfolio. These funds may
also specialise in merger and acquisition (M&A)
operations, which entails taking positions in the
securities of the target company, and where
appropriate taking an opposite position in the
securities of the company initiating the
Directional strategies and tactical approaches differ
considerably from the theoretical framework
described above to outline the principles of
alternative investment management. Here,
positions are taken that either follow a market
trend or “lean against the wind”, depending on
whether the fund manager expects the trend to
continue or to turn around. By definition, these
positions are unhedged. Implemented by “global
macro” funds, these strategies are not based on a
technical financial market approach, but rather
on a financial and macroeconomic analysis of the
situation of a country, sector of activity or
economic area. Trading funds, “market timing”
and “trend followers” funds, futures funds and
Commodity Trading Advisers implement these
investment strategies. They are closer to
conventional “speculative” position-taking than
to other alternative investment strategies that
take advantage of valuation inefficiencies.
5The notion of arbitrage should not be interpreted here in the usual sense of “risk free” arbitrage, but rather as a position aiming to exploit a
“temporary” divergence between the movements of two assets that is not typical of their long-term relationship (see Part 3).
6Convertible bonds are frequently undervalued in relation to their theoretical value. The most conventional arbitrage technique consists in taking
a long position in a convertible bond and a short position in the underlying stock, so as to exploit the price difference while remaining protected
against fluctuations in the stock itself.
eulavevitalerdnaegartibrA %63
seigetartslanoitceriD %03
latoT %001
Table 1
The alternative investment industry:
relative share of different investment styles
Challenges arising from alternative investment management
4 Banque de FranceFinancial Stability Review No. 3 November 2003
As different as they may be, the investment
styles described above all require the fund
manager to be extremely proactive and adopt
an opportunistic and discretionary approach.
These particularities, and the constraints they
impose on investors, especially in terms of the
liquidity of their investment, explain why the
alternative investment industry has developed
within the framework of ad hoc structures that
are not subject to prudential rules (risk division
ratios, strict rules covering the use of derivatives,
short selling, etc.) and transparency rules
(valuation) imposed on standard investment
vehicles, and are accessible only to a minority
of qualified investors. These investment funds
are commonly known as “hedge funds” or
“highly leveraged institutions” (HLI). However,
no statutory definition exists for these terms, nor
do market participants unanimously agree on
one 7. In very broad terms, from an operational
standpoint, hedge funds can be qualified as all
investment vehicles not bound by ordinary law
in terms of protection of savings and/or that are
liable to use a complete range of available
financial market instruments and techniques.
In the rest of this paper we will employ the terms
“hedge fund” and “alternative fund”
1|2 An evolving industry
Investment structures and customer bases:
from global macro funds
to alternative multi-management
Over the past 15 years, the hedge fund industry has
evolved significantly. In quantitative terms, despite
the fact that alternative investment funds still only
account for a small proportion of funds invested in
the asset management industry, the number of
alternative funds has increased dramatically,
growing by a factor of around five since 1988, to
exceed 7,000 units. In particular, over the same
period, the amounts under management look to have
increased by a factor of almost 20, to reach
USD 650 billion at end-2002.
7A document published in the light of American Supervising Authority, the Securities and Exchange Commission (SEC), hearings on hedge funds
in May 2003 offers fourteen different definitions.
Chart 1
Hedge funds: number of vehicles and assets under management
(USD billions)
Source: Van Hedge Fund Advisors International Inc.
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 20031988
Number of funds (left-hand scale)
Assets under management (right-hand scale)
Challenges arising from alternative investment management
Banque de FranceFinancial Stability Review No. 3 November 2003 5
This booming industry has also radically
Global macro funds, which symbolised the
development of this industry during the 1990s,
from the sterling crisis of 1992 to the Asian and
Russian crises of 1997-1998, no longer dominate
the universe of alternative funds. Strictly
speaking, these funds currently account for less
than 5% of assets under management (compared
with almost 70% at the start of the 1990s) within
an industry dominated by arbitrage funds and
characterised by small specialised structures:
according to a number of market sources, 80% of
hedge funds currently manage assets worth less
than USD 100 million.
Chart 2
Breakdown of assets under management
(by portfolio size, in USD millions)
(17%) > 500
Source: Banque de France estimates.
The customer base of hedge funds, initially
concentrated on wealthy private investors, in
particular in the United States, steadily
broadened, to encompass institutional investors
(pension funds, insurance corporations, etc.), and
later on all individual investors. The
development of multi-management and “funds
of funds”, and the popularisation of alternative
funds offering capital guarantees plainly
illustrate the current “democratisation” trend of
alternative investment products.
There are several reasons behind institutional and
individual investors’ recent enthusiasm for
alternative investment funds, some of which are
cyclical, and others more structural.
In addition to the decline in bond yields since
the start of the 1980s, equity markets have fallen
substantially over the past few years. The former
trend prompted investors to seek new sources of
high returns, while the latter highlighted the
relevance of approaches aiming to achieve
absolute performance, regardless of trends in the
underlying markets.
The propensity of traditional asset classes to
exhibit, particularly in bear markets, a high
correlation, and the limitations of international
diversification stemming from the increasing
globalisation of financial markets, have also
contributed to this trend.
The alternative investment industry has also
expanded as a result of “deviations” in traditional
management, in particular the practice of “closet
indexing” in benchmarked asset management.
Lastly, over the most recent period, the increase
in volatility in all markets may have contributed
to the more commonplace use of alternative
investment management.
Challenges arising from alternative investment management
6 Banque de FranceFinancial Stability Review No. 3 November 2003
2| Performance measures and alternative risk strategies
2|1 The difficulties entailed in
measuring risks in alternative
investment strategies
Alternative investment strategies are infinitely more
complex than those implemented by traditional
funds, such as buy-and-hold strategies, as, in addition
to identifying the markets in which funds take
positions (i.e. the location factor), it is also necessary
to calculate their net exposure and leverage (i.e. the
strategy factor). Measuring their performance is
rendered all the more difficult.
Traditional measurement tools ...
There are two main approaches used to measure
performance: “absolute” measures, which are
independent of any benchmark, and “relative”
measures, which compare an asset’s performance
to that of a benchmark portfolio. The first category
includes indicators such as the Sharpe ratio and the
Treynor ratio. These measures consist in calculating
the excess return of an asset minus the risk-free
rate over the risk indicator. The latter can be defined
as the standard deviation of the asset’s return series
(i.e. return volatility, see the Sharpe ratio), or the
sensitivity of the asset to market movements
(i.e. beta, see the Treynor ratio), or a measure of
maximum loss (i.e. Value-at-Risk or VaR 8). Among
the tools for measuring the relative performance
adjusted for risk, Jensen’s alpha (1968), the best
known, is defined as the average return on a
portfolio over and above that predicted by the
market model, given the portfolio’s beta and the
average market return.
... which are poorly suited to measuring
the specific nature of hedge fund performance
Exposure to multiple risk factors
Given that they are free to invest dynamically in a
large range of assets, hedge funds are exposed to
multiple risk factors (market, volatility, credit,
liquidity, etc.). Consequently, even though some
strategies are non-directional, the risk-free rate
cannot be an appropriate benchmark and the
absolute performance measures discussed above are
not suitable for measuring hedge fund performance.
Among the relative performance measures, only
those using a benchmark that takes account of all
the sources of risk to which hedge funds are exposed
are apposite for measuring the risk-adjusted
performance of alternative strategies. Many
difficulties are still encountered in measuring these
risk factors and apprehending their interdependence
(e.g. liquidity and credit risks). This explains why
only a few strategies have, to date, given rise to
relatively robust models (such as trend-following,
merger arbitrage, and more recently bond
Dynamic and non-linear exposures to risk factors
Most performance evaluation methods currently
implemented use one-factor linear or multifactorial
models 9. The effectiveness of these models
depends, inter alia, on the linearity of the
relationships between the dependent variable and
the explanatory variables. Unfortunately, three
factors contribute to the non-linearity of the
exposure of hedge fund performance to the
different risk factors.
A number of risk factors underlie each alternative
strategy. For a given strategy, the best funds are
those which succeed in correctly over- or
underweighting their exposure to different risk
factors. Because they use this “tactical factor
allocation” strategy, and “market timing” or “risk
factor timing”, their exposure to the different risk
factors changes over time.
Moreover, given that markets are relatively
efficient, there are not an infinite amount of
arbitrage opportunities. In order to maintain their
performance, some hedge funds therefore tend
to seize any opportunities that arise, even if it
makes them deviate from the strategy they claim
to implement. These one-off changes in style,
known as “tactical style allocation” by fund
managers and “style drift” (see Lhabitant, 2001)
by investors, also result in variations in exposures
to risk factors. The exposure of hedge funds to
risk factors is hence doubly dynamic.
8VaR expresses the portfolio risk as the maximum amount of loss that can be incurred for a set confidence threshold at a given time horizon.
9CAPM, a three-factor model developed by Fama and French; a four-state model by Carhart; Arbitrage Pricing Theory (APT), etc.
Challenges arising from alternative investment management
Banque de FranceFinancial Stability Review No. 3 November 2003 7
Box 1
Traditional performance measures
Absolute measures
Sharpe ratio :
E (R
) – R
σ (R
; Treynor ratio :
is the expected portfolio return,
its volatility,
the risk-free rate, and
P the sensitivity
of the portfolio return to market variations.
Relative measures
Jensen’s alpha is calculated by the following regression:
– R
= α
– R
) + ε
Mt are the return on the portfolio, the risk-free asset and the market at date t,
P the sensitivity
of the portfolio return to market variations and
an error term.
Contrary to the Sharpe and Treynor ratios, Jensen’s measure contains a benchmark. However, Jensen’s alpha
cannot be used to compare portfolios with different risks, as the value of alpha is proportionate to the level of risk
To solve this problem, Modigliani and Modigliani (1997) proposed M 2 or Risk adjusted performance (RAP). This
measure evaluates the performance adjusted for the risk of a portfolio against the market benchmark, expressed
as the return per unit of risk:
= (E (R
) – R
) + R
σ (R
σ (R
E (R
is the expected portfolio return,
its volatility,
that of the market and
the risk-free
Lobosco (1999) then further developed RAP to SRAP, to take account of the effect of management style on an
asset’s performance. The SRAP is defined as the difference between the asset’s RAP and the benchmark’s RAP,
representative of the asset’s management style.
Lastly, in the Sortino ratio, which is very close to the Sharpe ratio, the risk-free rate is replaced by the benchmark
return (or the Minimum acceptable return MAR) and the standard deviation by the square root of the
semi-variance. This indicator can then be used to evaluate the performance of an asset whose return distribution
function is not symmetrical.
Sortino ratio :
is the expected portfolio return, MAR is the Minimum acceptable return and
the portfolio
return at date t.
E (R
) – MAR
– MAR)
t = 0
E (R
) – R
Challenges arising from alternative investment management
8 Banque de FranceFinancial Stability Review No. 3 November 2003
The asset portfolios held by hedge funds (in
particular derivatives) are themselves sources of
non-linear exposures to the different risk factors.
The remuneration system of hedge funds
comprises a fixed part (management fees)
and a variable part (incentive fees). This
remuneration is therefore asymmetrical as the
return pattern of the variable part is similar to
that of a call option on the fund’s performance 10.
Since the performance of hedge funds is
reported net-of-fees, this introduces de facto a
non-linear component.
Traditional indicators for measuring absolute
performance assume that the (total or systematic)
risk is constant over the whole analysis period.
Likewise, standard mono- and multi-factorial
models do not take into account risk factor
exposure dynamics, as the stability of the
coefficients is one of the central assumptions of
these models. They simply measure the average
exposure to different risk factors over the analysis
period. Naturally, this distorts the evaluation of
hedge funds’ risk-adjusted performance.
Extreme risks
Most traditional tools for measuring performance
and risks assume the normality of the distribution
of returns on the evaluated asset. The asset’s risk is
therefore characterised by its volatility (i.e. the
standard deviation). However, many studies have
shown the significance of third-order moments
(asymmetry coefficient) and fourth-order moments
(flattening coefficient) of the distribution functions
of hedge fund returns. Hedge fund performance
therefore cannot be analysed using the mean-
variance framework.
Taking into account exceptional events exacerbates
the statistical estimation problems that arise from
using VaR 11 . To overcome these difficulties, investors
and fund managers have implemented some
interesting solutions: stress tests, scenario analyses,
more complex modelling of distribution tails with
extreme value theory. Nevertheless, although they
are theoretically more robust, there are still practical
difficulties in applying them due to gaps in
knowledge about the underlying factorial structure
of the different alternative strategies (above all
scenario analyses and stress tests) and the limited
number of observations (especially in the case of
the “block maxima” method and the peaks-over-
threshold method, or more simply for expansions
of Cornish-Fisher-type – VaR approach) 12.
Relative measures of the performance
of alternative strategies
Despite the fact that they are unsuitable for
evaluating hedge fund performance, absolute
measures of performance have been used in many
studies. While there is no doubt that the evaluation
of the performance of alternative strategies needs
to take account of the risk factors to which they are
exposed, the best method for doing this has yet to be
determined. Where some authors used a single-factor
model, others employed multi-factorial models to
take better account of the diversity of risk sources.
For all that, the factors in hedge fund returns remain
difficult to identify; the risk being that incorrect
factors may be included or correct ones left out.
Implicit factorial analysis
Hedge funds do not generate the majority of their
returns from allocation between different asset
classes, as is the case for traditional funds, but from
the dynamic strategies implemented by their
managers. However, Sharpe’s model for analysing
management styles does not take into account the
active component of hedge fund returns. The results
yielded are therefore not very significant when
applying the model to hedge funds. According to the
study by Fung and Hsieh (1997) the explanatory
power (R 2) of Sharpe’s model is indeed below 25%
for 48% of hedge funds, whereas it exceeds 75% for
47% of mutual funds). Therefore, in order to improve
the explanatory power, it is necessary to integrate
the factors reflecting the specific nature of hedge
fund strategies (i.e. trading factors). The best solution
for identifying these, without being exposed to a high
degree of model risk, is to carry out a factorial
analysis of hedge fund returns in order to determine
the predominant styles. This method is based on
the assumption that managers investing in the same
asset classes and using the same management
strategies must generate correlated returns. This
analysis method can be used to explain the variations
in hedge fund returns, but it does not provide a clear
10 The strike price is therefore equal to the “hurdle rate”, i.e. the return above which the fund is paid incentive fees.
11 Historical VaR calculations require a large number of variables to obtain a significant event sample; VaR calculations based on a
Monte Carlo approach are complex and cumbersome; parametric analyses are oversimplified.
12 See Jorion (2001) for further details on VaR and Lhabitant (2003) for a presentation of extreme value theory.
Challenges arising from alternative investment management
Banque de FranceFinancial Stability Review No. 3 November 2003 9
view of the dynamics of hedge fund returns over
time. Furthermore, a significant part of the
variation in hedge fund returns in the sample
studied is not explained by the main factors
identified (e.g. over 50% according to the study
by Fung and Hsieh, 1997).
Analysis using a linear model with non-linear regressors
In order to compensate for the lack of traditional
linear factorial models for analysing hedge fund
performance, models have been developed in the
literature that make it possible to take account of
the non-linearity of these funds’ returns. New
regressors (or explanatory variables) are used that
have a non-linear exposure to traditional asset
classes, so as to approximate dynamic management
strategies in a linear regression. Option portfolios or
hedge fund indices are natural candidates for these
new regressors.
Given that the management strategies implemented
by hedge fund managers are not known with
precision, some authors have attempted to describe
them using simple optional strategies. By
implementing models comprising call and put option
portfolios of certain location factors (equity indices,
bond indices,), a default factor (i.e. credit spread),
Fama and French factors (size, value/growth) and a
Carhart factor (momentum), they managed to
explain a significant proportion of the variability in
hedge fund returns over time 13.
Another possibility would be to use hedge fund
indices. This approach is based on an extension of
the style analysis model developed by Sharpe (1992)
for traditional funds. By extending this model it is
possible to take into account specific hedge fund
strategies, i.e. the use of short-selling and leverage.
With this choice of factors we can eliminate
non-linearity problems, which are taken into
account in the indices themselves. The accuracy of
results depends directly on the quality of the hedge
fund indices used. It is therefore highly
recommended to use style indices that are both
representative and with little bias (see the box on
EDHEC indices below). This model can be used to
compare the performance of hedge funds with an
appropriate benchmark, without knowing with
precision the management strategy applied by the
hedge fund. This model is simple to implement and
uses only hedge fund returns. Its main weakness is
that it assumes coefficients allocated to the different
indices are constant over the whole analysis period.
It is therefore difficult to capture the dynamics of
exposures to risk factors.
Non-linear analysis: payoff distribution pricing model
A final approach involves using a non-linear model
to explain hedge fund returns. In order to correctly
evaluate the performance of portfolios with
non-normal return distributions and a non-zero
asymmetry coefficient, the distribution as a whole
must be considered. Ideally, this should be done
without specific assumptions regarding the
distribution pattern. This approach therefore
proposes an efficiency test whose theoretical
foundations are based on the payoff distribution
pricing model (PDPM) by Dybvig (1988a, 1988b).
Dybvig’s model attributes a price to a given
consumption distribution function (i.e. the price is
equal to the cost of the least expensive portfolio
generating the consumption function). The
difference between the cost of the investor’s real
portfolio and the cost of the least expensive portfolio
generating the same consumption function seems
the natural measure in monetary units of the
efficiency loss. The advantage of this evaluation
model is that it does not require assumptions on the
distribution of the returns of the funds considered.
The results of an empirical study carried out
on a sample of 1,500 hedge funds obtained
from the MAR/CISDM database 14 and using all
the above-mentioned performance measurement
models show that traditional mono- or
multi-factorial models find significantly positive
alphas for hedge funds. However, when the whole
distribution of returns is taken into account or only
implicit factors are included in the model, hedge
funds no longer, on average, have significantly
positive alphas. All in all, the alpha of a hedge fund
may have a dispersion of over 40% between the
different methods! This reminds investors that,
above and beyond the financial and operational risks
to which hedge funds are exposed, model risks must
also be taken into account.
13 This approach is perfectly illustrated in studies by Fung and Hsieh (1997) for trend following strategies, Mitchell and Pulvino (2001) for merger
arbitrage strategies, Fung and Hsieh (2002c) for bond strategies and Agarwal and Naik (2003).
14 Amenc et al. (2003a).
Challenges arising from alternative investment management
10 Banque de FranceFinancial Stability Review No. 3 November 2003
2|2 The difficulties
of benchmarking in alternative
investment management
Given that the risk-free rate is not a suitable
benchmark for all types of hedge funds, an
appropriate benchmark remains to be determined.
It appears that the alternative investment industry
is shifting from an absolute return approach to a
relative return approach. The principle involves
comparing a given fund’s returns with those of a
portfolio of a fund implementing the same strategy
(peer benchmarking), or with those of a
benchmarking index. Already difficult in the
traditional universe, compiling appropriate indices
is much more complicated for the alternative
investment industry, due to problems associated
with both representativity and the purity
(i.e. homogeneity) of data.
Impact of biases in databases
The use of a specific data sample, from a universe of
hedge funds that cannot be observed as a whole,
introduces a bias in the measure of performance.
There are essentially three sources of difference
between the performance of hedge funds calculated
using this type of database and the performance of
hedge funds as a whole. These different biases, which
are described in Fung and Hsieh (2000 and 2002a),
are: the survivorship bias, the selection bias and the
instant history bias.
The survivorship bias occurs because poor managers
leave the industry, and good managers remain.
Furthermore, the funds in databases tend to be
those whose return exceeds the average return for
all hedge funds, as databases only contain the
returns of good managers, or at least those of
managers present at the time of measuring. The
standard procedure for measuring the survivorship
bias is described by Malkiel (1995). It involves
determining the difference over the period under
review between the average return for all hedge
funds and the average return from surviving funds.
As the composition of indices varies significantly
between the different data providers, the
survivorship bias will not have the same effect on
the different hedge fund indices.
The selection bias results from the fact that the
selection criteria of databases may differ
significantly, and the data provided by these
databases are not necessarily representative of the
same management universe and of the real hedge
fund universe. Yet again, the impact of this bias
will depend on the databases’ selection criteria.
The selection bias will therefore not be the same
for all indices.
Furthermore, data on hedge funds are not easily
accessible. Reporting the performance of an
alternative fund in one of the competing databases
is purely voluntary and only some funds choose to
do so. This results in a “self reporting” bias. As funds
not reporting their performance in any database are
by definition unobservable 15, it is impossible to
evaluate the impact of this bias.
The instant history bias (see Park, 1995) results
from the difference in the dates when hedge fund
data are entered in databases. Yet, when a fund is
added to a database, some or all of its history must
be back filled. This can be done by extrapolating
the fund’s recent performance data. In general, as
newly-entered funds are likely to perform well, it
is probable that their average performance over
the incubation period will be better than that of
funds that have been in the database for a long
time (upward bias on returns of newly-entered
funds). If funds are not entered on the same date
in two different databases, they will not be equally
exposed to this bias.
The heterogeneity of hedge fund indices
Competing indices are compiled using different
databases and a variety of construction methods.
Consequently, they are not affected in the same
way by the measurement biases described above.
Differences in performance therefore arise.
15 Some funds chose not to report their performance because it is not satisfactory, and others because they have already reached critical size.
saiBnnamzteoGdnanworB,kraP )9991( heisHdnagnuF )0002(
latoT %5.4 %4.4
Table 2
Survivorship bias and selection bias
in hedge fund returns
Sources: authors quoted
Challenges arising from alternative investment management
Banque de FranceFinancial Stability Review No. 3 November 2003 11
Major disparities in performance are often observed
in the various competing indices for a given
management style. This phenomenon is particularly
marked in crisis periods, in particular between
August 1998 and October 1998 (see Table 2). Data
provided by the different index providers differ
substantially: by over 20% in the case of the
performance of the Zurich and EACM (Evaluation
Associates Capital Markets) long/short equity
indices in February 2000 (see Amenc and Martellini,
2003). An analysis of mean and median correlations
between the performance of different competing
indices confirms this lack of homogeneity. The
mean correlation between competing indices on a
given type of strategy (equity market neutral: 0.43,
long/short equity: 0.46) may be less than 0.5. The
increasing number of index providers and
construction methods result in a greater
heterogeneity of data. If we consider that the
heterogeneity indicator HI=1 – mean correlation 16,
it is obvious that hedge fund indices do not provide
the representativity conditions necessary to offer
investors a homogenous vision of alternative funds.
In order to document the heterogeneity of
different indices we can also consider the
differences in exposure to the main risk sources
of these strategies. The results given in Table 4
speak for themselves. Once again, we clearly
observe that the different competing indices are
very heterogeneous. In view of the different
exposures to the various risk factors, the results
will differ according to the competing index chosen
as the benchmark.
16 Perfect heterogeneity of indices is expressed as HI=1.
Table 4
Sensitivity to the main risk factors – the fixed income strategy
(from January 1998 to December 2000)
dexiF emocni egartibra
tekraM ksir 0
ytilitaloV ksir 0000
etartseretnI ksir 000 0
ehtfoepolS evrucdleiy egnahcxE ksiretar seitidommoC ksir 0000 tiderC ksir 0ytidiuqiL ksir 0
RFH 61.0- 41.0 52.0 91.0 75.0 70.0 42.0- 81.0-
eessenneH 73.0 73.0- 60.0 91.0 62.0 21.0 22.0- 21.0-
NB: Market risk: relative variations in the Standard and Poor’s 500 Price Index; Volatility risk: relative price variations in the VIX contract; Interest rate risk: variations
in the three-month Treasury bill yield; Exchange rate risk: USD exchange rate against a basket of foreign currencies; Commodities risk: relative price variations in
a barrel of crude oil; Liquidity risk: changes in the volumes of securities traded on the New York Stock Exchange (NYSE); Credit risk: relative changes in the spread
between yields on bonds rated BAA and AAA by Moody’s; slope of the yield curve: spread between the yield on a 30-year bond and a three-month T-Bill.
Source: Armenc and Martellini (2003).
selytstnemtsevnI )secidnidnasetadhtiw(secnereffidmumixaM
egartibraelbitrevnoC 0%67.4 )%80.0(eessenneH/)%86,4-(BFSC:8991rebotcO
seitirucesdessertsiD %19.6 )%22.8(hcirüZ/)%13.1(MCAE:0002yraurbeF
lartuentekramytiuqE %00.5 )%02.5(egdeHnaV/)%02.0(eessenneH:9991rebmeceD
egartibraemocnidexiF %03.11 )%02.0(egdeHnaV/)%01.11-(teNFH:8991rebotcO
sdnuffosdnuF 0%10.8)%24.01(tsevtlA/)%14.2(RAM:9991rebmeceD
orcamlabolG %71.41 )%26.2(tsevtlA/)%55.11-(BFSC:8991rebotcO
egartibraregreM %78.1 )%59.2(tsevtlA/)%80.1(teNFH:8991yraurbeF
eulavevitaleR%74.01 )%04.4(egdeHnaV/)%70.6-(MCAE:8991rebmetpeS
gnillestrohS %02.12 /)%03.42-(egdeHnaV:0002yraurbeF)%01.3-(MCAE
Table 3
Differences in maximum performance observed beween representative indices of a given style
(from January 1998 to December 2000)
Challenges arising from alternative investment management
12 Banque de FranceFinancial Stability Review No. 3 November 2003
Just as in the traditional management universe, the
different indices published are neither collectively
exhaustive nor mutually exclusive. However, in the case
of the alternative investment universe, the lack of
Box 2
The EDHEC alternative indices
As no consensus has been reached on the best methods for classifying funds or constructing alternative indices,
it is not possible to determine objectively which of the existing indices is the most representative. Rather than
arbitrarily selecting one, EDHEC suggests choosing all of them and then combining them in such as way as to
obtain a relevant representation of the performance of alternative strategies. By selecting all the indices, it is
possible to aggregate the underlying databases and thus obtain a more representative index. In order to achieve
the optimum solution, we can then use principal component analysis (PCA). It should be recalled that the first
axis of the PCA gives the linear combination of the original variables (e.g. the series of returns of competing
indices) that make it possible to capture the greatest amount of data contained in the variance/co-variance
matrix of the original variables. Then, it is just a matter of selecting the first axis of the PCA and standardising
the weights to obtain an index of indices that maximises the degree of representativity. The weightings of indices
of indices are then recalculated every three months in order to maintain the quality of representativity. In order
to ensure that the method is efficient, indices of indices are compared to the portfolios of funds consisting of a
very large number of funds (i.e. highly representative). The higher the correlation coefficient with the portfolios
of funds, the more representative the index. Table 5 compares the correlation coefficients obtained by the EDHEC
indices (i.e. indices of indices) with the average correlation coefficients obtained by the indices available in the
market. Note that the database used for this test contains 7,422 funds (including 2,317 not recorded in any
database) 1. Indices of indices are systematically more representative than the average of the indices of which
they consist.
1We would like to greatly thank François-Serge Lhabitant for providing us with the series of portfolio returns obtained from a database of
7,422 hedge funds.
selytstnemtsevnICEHDE secidni gnitepmoC secidni
Table 5
The degree of representativity
Principal Component Analysis can be used to obtain
the linear combination of the competing indices that
make it possible to capture the greatest amount of
data contained in these different indices. As the table
opposite shows, this results firstly in a maximisation
of the degree of representativity. Secondly, and of
equal importance, indices of indices are, by
construction, systematically less biased than the
indices of which they consist due to the fact that
competing indices are not all affected in the same way
by the performance measurement bias. Therefore, the
maximisation of the explained variance implicitly
leads to a minimisation of the biases. This property
of indices of indices is useful for evaluating the
performance of alternative strategies.
regulation and the associated lack of transparency
exacerbate this problem considerably. It is therefore
crucial to construct style indices that meet sector
professionals’ transparency and reliability requirements.
Challenges arising from alternative investment management
Banque de FranceFinancial Stability Review No. 3 November 2003 13
3| Alternative investment management
from the financial stability perspective
From the financial stability perspective, the
development of the alternative investment industry
raises two sets of questions. First, to what extent does
this management approach and the techniques
involved impact on the functioning and dynamics
of financial markets? Second, in view of the specific
risks associated with the alternative investment
industry, how should roles be divided between
market participants and supervisory authorities for
the oversight of this activity and its development?
3|1 Contribution to the functioning
of financial markets
Given the diversity of alternative investment
strategies, it is not possible to unequivocally assess
the impact of the alternative investment industry
on the functioning of financial markets. An approach
by management style and an analysis of techniques
used by alternative fund managers prove to be more
fruitful, and also make it possible to pinpoint the
specific risk factors associated with alternative
investment management.
More complete financial markets
A detailed examination of the functioning of financial
markets reveals that “pockets” of inefficiency, as
defined in financial theory, i.e. situations in which
asset prices do not reflect all available fundamental
data, are likely to develop. These anomalies lead
in fine to asset price misalignments. A number of
investment strategies implemented by alternative
investment firms, in particular non-directional
strategies, aim to exploit these situations 17:
by providing market liquidity when there appears to
be a lack. In general, given that alternative funds
are not subject to the constraints governing most
other market participants, they can deal with
market shocks better, and therefore provide
liquidity when markets are unbalanced. On
“distressed” debt markets, or high-yield bond
markets, which are narrow and fragile markets
because they do not have a broad natural investor
base, demand from alternative funds may
contribute to limiting excessive price
misalignments. Likewise, the spectacular
development of the convertible bond market in
recent years is largely due to the capacity of funds
specialising in this type of arbitrage to absorb an
increasing supply of paper;
by actively participating in the financial asset price
discovery process. “Long/short” strategies and
arbitrage strategies contribute to ensuring the
overall consistency of financial asset prices.
Alternative funds’ strategies are very different in
this area from those of traditional fund managers
and may also have, ceteris paribus, a greater
impact: on the one hand, in their allocation
decisions, they do not face the constraint of not
being allowed to deviate excessively from a
benchmark portfolio and, on the other, they can
short sell securities deemed to be overvalued.
Fuelling destabilising dynamics?
The fear that hedge funds’ strategies may be fuelling
destabilising market dynamics would appear to
contradict the “merits” often attributed to them,
i.e. that they are thought to reduce volatility and
foster greater financial market stability. The
continuing debates over short selling, and its use
by hedge funds, illustrate the complexity of this
issue. There is no doubt that short selling may, in
the context of fragile markets, amplify an imbalance
and facilitate the development of cumulative
depreciation 18. However, this technique, which is not
only used for speculating on the fall in the price of a
specific security, is also implemented by other
players. Furthermore, it is often combined with long
positions (i.e. in long/short strategies and arbitrage
strategies). Lastly, it should be noted that short
selling can be replicated synthetically using standard
market instruments such as swaps.
17 It should be noted that the operating methods of these funds do not differ fundamentally from those of the proprietary traders of commercial and
investment banks.
18 The same argument can be put forward for leveraged securities purchases and the use of derivatives for leveraging a long position, which fuel
cumulative appreciation.
Challenges arising from alternative investment management
14 Banque de FranceFinancial Stability Review No. 3 November 2003
More generally, most academic studies carried out
on this subject do not attribute an intrinsically
destabilising role to alternative investment strategies:
it does not appear that hedge funds’ positions on
markets systematically fuel deviations in asset price
from “fundamentals” 19, nor do they seem to lead to
herd or mimetic behaviour among other market
participants or “positive feedback trading” 20.
This diagnosis can be refined by breaking down
the impact of alternative strategies according to:
type of management style: directional strategies,
in particular “trend following” strategies, are by
nature more prone to exacerbate sharp price
movements and amplify the short-term trends
prevailing on markets. The fact that these
strategies rely to a large extent on leverage
aggravates this problem (see below);
the overall configuration of the markets: when
markets are stressed, they are more likely to
become unbalanced by participants that take
speculative positions, that use a large array of
instruments or that can more easily make use of
type of underlying market: the successive exchange
rate crises of the 1990s, in both developed
countries (ERM) and emerging countries (Asia
and Latin America) have demonstrated that in
some circumstances a speculative attack on a
currency may be self-fulfilling if the market
becomes convinced that, in the short run, it will
be too costly for the authorities to defend the
exchange rate, even if it had initially been
sustainable. These episodes have also showed
that, even though it is traditionally considered to
be one of the most deep and liquid markets, the
foreign exchange market is not immune to such
Irrespective of the potential destabilising effects of
alternative investment strategies, it should be
recalled that the investment mechanisms used in
traditional assets management may also have a
destabilising impact (short-termism, strict
benchmarking, etc.) 21.
3|2 Exposure to specific risks
Alternative investment management is often
presented as offering investors protection against
adverse market movements (i.e. against the risk of a
fall in prices). This argument should however be
qualified. Furthermore, a possible protection against
market risk does not mean that no risks exist, but
rather that specific risks are present.
Market risk is not systematically absent
from alternative strategies
Aside from directional strategies, which, clearly, are
fully exposed to this form of risk, it is interesting to
note that alternative funds exhibit not only very
different correlation levels with traditional asset
classes (equities and bonds), but also that these
correlation levels change in line with the direction
of markets: studies on conditional correlations have
demonstrated that the correlation levels with the
stock market of many alternative strategies increased
during periods of sharp falls in the market, i.e. exactly
when decorrelation would be most useful.
Not only does the alternative investment industry face
persistent market risk, but also more specific risks.
Operational risks
The strategies deployed by alternative fund
managers are often extremely complex, requiring the
extensive use of mathematical modelling and
sophisticated quantitative techniques. Model risk,
whether arising from the definition or the
assessment of the risks incurred (see Part 2), is of
particular relevance. While it is not the only factor
responsible for the spectacular collapse of LTCM
(Long Term Capital Management) in autumn 1998,
the unsuitability of risk measurement tools
(inadequate calibration of the parameters used to
calculate VaR, insufficient records of data used and
non-stationarity of series) is one of the main reasons
behind this failure 22. The fact that the relative value
strategies implemented by hedge funds have to be
analysed as statistical convergence strategies rather
than arbitrages in the strict sense of the term makes
it is even more important to focus on model risk.
19 Fung and Hsieh (2000): “Measuring the Market Impact of Hedge Funds”, Journal of Empirical Finance.
20 Eichengreen (1999).
21 Committee on the Global Financial System (CGFS) (2003).
22 Jorion (2000).
Challenges arising from alternative investment management
Banque de FranceFinancial Stability Review No. 3 November 2003 15
“Manager risk” may also be considered, among
operational risks, as a specific risk that includes
components such as the quality of the decision-making
process and the actual expertise of the fund manager,
which is a crucial factor in the alternative universe.
From liquidity risk to risks associated with leverage
This risk, while it is not specific to alternative funds,
takes on a particular relevance and form in the
context of alternative investment. It is especially
relevant because these funds invest in assets whose
secondary markets lack depth and liquidity 23. It is
also significant because alternative vehicles
frequently finance their positions using borrowed
funds or, to a greater degree, using leverage. Leverage
takes many forms: bank loans, derivatives or repo
transactions, short selling techniques and buying on
margin. Yet again, we should stress that alternative
funds are not the only market participants that use
these techniques or leverage. What sets them apart
is that they are not constrained by regulatory limits
governing this practice. All other things being equal,
the use of leverage exacerbates the different risks
incurred by funds and reduces their resilience to deal
with unforeseen shocks. In particular, it increases
market risk, via the collateralisation mechanisms
associated with debt financing and derivatives
transactions: a fall in the price of underlying assets
will trigger margin calls and/or the rapid liquidation,
under probably unfavourable conditions, of positions
refinanced in this way. This is one of alternative
funds’ factors of fragility, and a potential source of
contagion of the difficulties of some of its participants
to the market as a whole.
It is nevertheless difficult to accurately assess the scale
of the use of leverage by alternative funds: they do
not have any mandatory disclosure requirements in
this area, and the available data is patchy as it only
refers to leverage recorded in the balance sheet.
According to available market data 24, it appears that
almost three-quarters (74%) of alternative funds may
habitually use leverage in their transactions (funds
specialising in defaulted bonds make the least use,
with only 52% resorting to this technique, while macro
and arbitrage funds are the main users with 89% and
82% respectively). The amount of leverage remains
limited, with average leverage ratios of around 2 to 1
for hedge funds as a whole, and only exceeds this value
for 30% of funds. This is a far cry from the extreme
values registered by LTCM (50 to 1).
3|3 The regulatory framework
of the alternative investment
industry: a division of roles
between market participants
and supervisory authorities
The alternative investment industry has developed
in an environment very largely free from the
regulatory constraints that limit “traditional” market
participants. It could also be said that the industry
grew as a reaction to the restrictions that these
regulations imposed on its participants and to
exploit the effects of these limitations.
Developments to date have shown that alternative
investment management, according to the form it
takes, may contribute positively, to differing
degrees, to the functioning of financial markets. We
cannot however conclude that the development of
this industry should necessarily be unregulated.
Attention should be squarely focused on the
conditions in which the alternative investment
industry is developing for two main reasons.
The LTCM episode in 1998 illustrated how the
threat of a systemic crisis arising from the failure
of a key player, but whose importance on these
markets was only realised when it collapsed,
could weigh on financial markets.
The alternative investment industry is currently
opening up to new types of investors: on the one
hand, individual investors (in addition to its initial
wealth management and private banking clientele)
and, on the other, institutional investors, both of
whom the law sets out to protect.
The prevention of systemic risk and the protection
of savings are the two main reasons for
implementing regulations to frame the
development of the alternative investment
industry. As it stands today, this framework is built
around the close co-operation of market
participants and the public authorities, combining
the self-discipline of the former and the limited
regulatory constraints of the latter. In the medium
term, the more effective the implementation of
market discipline, the less need there will be to
extend this regulatory framework.
23 “Congestion risk” is a particular expression of this liquidity risk liable to affect the conduct of an investment strategy: a strategy that is profitable
will gain in popularity and attract new funds until the initial opportunities are exhausted for all those except participants willing to accept an
ever-increasing amount of risk.
24 Van Hedge Fund Advisors International.
Challenges arising from alternative investment management
16 Banque de FranceFinancial Stability Review No. 3 November 2003
Market discipline
Above and beyond the concerns that had already
been voiced by regulators, even before the collapse
of LTCM, this major shock gave rise to a salutary
realisation on the part of the hedge funds themselves,
the institutions financing them and investors, about
the need to substantially improve the operating
procedures of this type of activity and the
supervision to which it should be subject by their
counterparties as well as by their customers.
This realisation led to the development of codes of
good practices 25 that attempt to establish standards
for the industry as a whole. The recommendations
they propose target in particular the internal
organisation of funds (identifying responsibilities,
organising risk monitoring), the rules applicable to
the management of different risks (valuation methods
and resources to be implemented) and disclosure of
information (to investors, credit suppliers, and even
regulators). While these codes represent a significant
step forward in terms of the governance and
transparency of funds, they are not however legally
binding 26, and their effectiveness therefore remains
uncertain. While the Financial Stability Forum
subscribes to this notion of self-discipline, it
nevertheless regularly assesses, together with the
Joint Forum, the progress made in these areas, in
particular in the light of recommendations in the
Report of the Fisher Group on enhancing disclosure.
Aside from the funds themselves, their
counterparties and investors have a key role to play
in the implementation of efficient mechanisms of
discipline and self-regulation.
Credit institutions that finance alternative funds
are at the centre of the market’s self-regulation
mechanism. This is particularly the case for prime
brokers that provide a complete financial and
logistical service to hedge funds, from the
financing, executing and post-market management
of transactions (settlement, delivery versus
payment, accounts holding and custody services),
to the management of margin calls, organisation
of securities borrowing/lending, and even external
information and organisation of marketing
programmes. They are the only players to have a
complete view, in real time, of the situation of
hedge funds.
The opening up of the alternative investment
industry to institutional investors also transfers
to the latter a specific responsibility in terms of
monitoring. Because they themselves are subject
to strict reporting requirements and must follow
a rigorous investment process, these investors are
in a position to lend significant momentum to
increasing the transparency of the alternative
investment industry. There is no longer any
doubt as to the beneficial effects of the ”discipline
of transparency”. As regards hedge funds, we shall
confine ourselves to the results of a recent study
analysing the reported performance of these
funds in the light of their accounts auditing
practices: alternative funds are not usually
subject to disclosure and auditing requirements.
Their performance reporting, recorded in a
number of specialised databases, is also carried
out on a voluntary basis. A study by Bing Liang
highlights the fact that the differences in
performance of the same fund from one database
to the next (in itself an anomaly) are less
significant for audited funds than for others. The
paper also shows that there is a positive
relationship between the size of the fund and its
use of auditors, and that funds which make the
least use of leverage on the one hand, and those
with the largest investor bases on the other, are
most likely to have their accounts audited.
Investors and alternative fund managers have
started fruitful discussions that would be
beneficial to build upon.
A regulatory framework that remains limited
To date, intervention by supervisory authorities has
remained fairly limited. Banking supervisors have
also favoured an indirect monitoring of hedge funds
by stressing, on the one hand, the principles of sound
risk management that should guide relationships
between credit institutions and their counterparties
of this type, and, on the other hand, by enhancing
where appropriate the disclosure of their exposures
requested from credit institutions: in France, for
example, for poorly-rated or unrated counterparties,
regulations on reporting large exposures have been
modified, with credit institutions now being required
to report exposures in gross rather than net terms.
25 “Sound Practices for Hedge Fund Managers” (February 2000); “Guide to Sound Practices for European Hedge Fund Managers” (August 2002).
26 It should be noted however that in France, the guide to sound professional practices currently being drawn up by the Association française de
gestion shall have, after approval by the Commission des opérations de bourse (COB), the status of professional standard.
Challenges arising from alternative investment management
Banque de FranceFinancial Stability Review No. 3 November 2003 17
Market supervisors, for their part, have focused on
defining (restrictively) the conditions for the selling
of investment products offered by alternative
investment funds, with a view to protecting
individual investors.
In this exercise, market authorities must take into
account individual investors’ growing appetite for
these types of funds as well as the increasingly
diversified range of alternative products offered by
multi-management or structured investment funds.
Without pronouncing on the outcome of these
deliberations by the market authorities in different
countries, they appear, at present, to be opting for a
change in the access conditions to these products,
and a clarification of the regulatory regime, in order
to prevent in particular an uncontrollable off-shore
development of these products. In practice, these
regulatory developments result in promoting the
indirect access to alternative products, through
multi-management and alternative funds of funds.
The development of the multi-management industry
In brief, multi-management gives investors the
opportunity to invest in a selection of alternative
funds offering different risk profiles rather than one
specific alternative fund with a particular risk
profile. Alternative funds of funds are the most
dynamic segment of the alternative investment
industry: they currently account for USD 200 billion
in assets under management, i.e. more than a third
of the assets managed by this industry. Funds of
funds have significant advantages for “new”
alternative investors given that they select
individual funds from within an expanding range
of products, while taking charge of monitoring
performance and relations with alternative fund
managers. These funds offer a diversification of risk
between several management styles, and also
reduce the risk of loss associated with the high
hedge fund mortality rate. The multi-management
approach shares some common characteristics with
the traditional management universe; using a
multi-fund manager is a way of managing the
agency and asymmetry of information problems
that underlie the development of the asset
management industry. When selecting funds the
multi-management fund manager plays a similar
role to that of the traditional fund manager in
his/her asset allocation decisions.
Technically, therefore, funds of funds appear to be
one of the favoured vehicles for the large-scale
development of the alternative investment industry.
In a number of countries, regulatory developments
also lend impetus to this trend.
Regulatory developments
The notion of qualified investor is used in most
jurisdictions to restrict access to alternative products
to investors considered capable of assessing the risks
inherent in this management approach and bearing
the potential losses. This restriction may take the
form of a minimum subscription requirement (as is
the case most frequently) and/or a minimum income
or wealth requirement.
These types of restrictions exist both on direct access
to alternative funds and on investment in funds of
funds. The more recent regulations on access to
funds of funds are generally much less restrictive
than those on direct access to individual funds. In
France, for example, access to organismes de
placement collectif en valeurs mobilières (OPCVM) à
procédure allégée, the closest asset management
vehicle to what is usually understood by a hedge
fund, is reserved for qualified investors, with a
minimum subscription requirement of EUR 500,000.
“Contractual funds” 27, created under the financial
security bill, will in due course become part of
this range of funds. However, new regulations on
multi-management funds (see Box 3), set the
minimum subscription requirement at EUR 10,000.
In Italy, direct access to fondi speculativi, also
restricted to certain investor categories, is subject
to an initial subscription of EUR 500,000, and
EUR 25,000 for funds of funds, accessible to
individual investors. In Germany, a bill currently
under review would restrict direct access to hedge
funds to institutional investors, and give individual
investors the possibility to invest in funds of funds.
These rules are equivalent to those in place in the
United States, where investors are subject to an
initial subscription of USD 25,000 for accessing
multi-management investment products, and
USD 200,000 (or net wealth of USD 1 million) for
hedge funds stricto sensu. In the United Kingdom,
alternative funds of funds can be marketed without
restriction to individual investors as long as they
are listed, but can only invest in funds actually
approved by the Financial Services Authority (FSA).
27 These funds, restricted to qualified investors, will operate without an authorisation from the supervisory authority (on the basis of simple
declaration), and will not be subject to any constraints in terms of division of risk. Nor will they face constraints in terms of asset types. They will
be free to establish the frequency and method of calculating net asset value.
Challenges arising from alternative investment management
18 Banque de FranceFinancial Stability Review No. 3 November 2003
Box 3
Regulating alternative multi-management investments
(COB press release of 3 April 2003)
In France, management strategies seeking to deliver absolute returns, uncorrelated with a benchmark index have
developed only marginally. They are not widely distributed and account for limited amounts under management.
Amid difficult market conditions, demand for these products has grown, mirroring the international development
of hedge funds. Today, these management strategies are commonly referred to as “alternative”, although this term
has no standard international definition and its substance varies significantly.
In France, alternative investments primarily consist of alternative funds of funds, i.e. French funds invested in
offshore funds or French funds with a specialist bias, such as futures or options). It is therefore necessary to establish
a precise legal framework which can be applied to an activity that France has tolerated for almost a decade.
Investment management companies that choose French or foreign funds relying on complex management
techniques much follow due diligence procedures, which need to be formalised and included in a special programme
of operations (for discretionary management and collective investment fund management). Furthermore, investors
must be informed of the special characteristics of such products and techniques through the marketing programme
and appropriate informational materials. In order for them to make an informed decision about a particular
product, prospective fund subscribers and discretionary clients must be clearly informed by means of the
fund’s prospectus, discretionary mandate and any promotional literature, of the type of investment involved and
the specific risks inherent in it.
Following several months of industry-wide discussions, the COB recently adopted a series of positions (...) that
establishes a framework for contributing to the development of alternative investment activity and ensuring
proper security. The main rules are as follows 1.
Management companies managing a fund or a mandate invested in an alternative fund must update their
programme of operations accordingly (companies seeking authorisation to exercise this activity must submit
this amended programme beforehand).
General purpose funds with less than 10 per cent of their assets vested in alternative funds must update
their prospectus accordingly (and the management company must update its programme of operations).
General purpose funds with more than 10% of their assets invested in alternative funds must update their
prospectus. Management companies must supplement their programme of operations with a marketing
programme (for newly formed funds, these documents must be submitted to the COB as a prerequisite for
The COB will review programmes of operations to ensure that management companies have the necessary
skills and ressources to manage these products.
Existing regulations will be adapted to cover alternative investment funds — notably via the creation of a new
classification: “funds invested in alternative funds” — in 2003. This will be part of an overall regulatory overhaul
aimed at taking into account the issues dealt with by the COB (e.g. the working group on management fees and
charges, chaired by Philippe Adhémar) as well as new European directives.
In addition, discussions are still underway with the French Investment Management Association, AFG, with a
wiew to approving a code of professional conduct for market participants involved in alternative investment
These discussions will be extended, in collaboration with the industry as a whole, to determine the best
arrangements for implementing direct alternative strategies in funds organised under French law.
1The decision statement and special programme of operations are available on the COB website at
Challenges arising from alternative investment management
Banque de FranceFinancial Stability Review No. 3 November 2003 19
All in all, the alternative investment industry does not deserve to be excessively praised or maligned.
Looking to it to find the solution for the difficulties recently encountered in traditional asset management
would be just as simplistic as branding it a systematic factor of financial market destabilisation:
alternative investment can indeed be an instrument for risk diversification and, in this respect, a
useful tool in overall portfolio management, while contributing to greater market efficiency. But it
can also contribute to, or even amplify market imbalances. More than simply a new asset class,
alternative investment management should be seen for what it really is: an ensemble of unconventional
investment strategies, often very different from one another, involving traditional asset classes, and
carrying its own specific risks. The spectacular movements observed on financial markets over
recent years, and the disappointment of investors, have accelerated the recognition of the alternative
investment industry and just as spectacularly fuelled its development. This first hurdle has now
been overcome but many more challenges still face the alternative investment industry: how can a
niche activity become a mature industry without losing its particular features? How, over time, can
performance and standardisation be reconciled? In the long term, the answer to these questions will
depend, in particular, on the balance struck between market discipline and the regulatory framework,
as well as on the alternative investment industry’s capacity to open up and enhance transparency.
We have seen how crucial the complex issue of measuring performance and risk is in this area. It
will also depend on the developments that the “traditional” asset management universe will necessarily
undergo: if it evolves towards a “core-satellite model”, this would reinforce the position of alternative
investment management alongside a more systematic index-linked approach. Conversely, if traditional
fund managers return to a more pro-active approach, this would no doubt limit the alternative
sector’s growth margin.
A recent study by the FSA has shown that individual
investors have little appetite for direct access to
hedge fund products.
As the new regulatory regime for alternative
multi-management in France shows (see Box 3), this
indirect access to alternative investment products
is often coupled with the implementation of a new
regulatory framework. The latter aims to ensure that
not only investors are protected (specific monitoring
of sales and marketing conditions, restricting access
to alternative investment products, banning direct
marketing of underlying funds and updating
prospectuses) but also, more broadly, that the
operational risk associated with this management
approach is prevented (management firms
concerned are required to submit a special
programme of operations to the COB for approval,
with a view, in particular, to ensuring the expertise
of management teams, the suitability of specific risk
control procedures, the existence of technical
resources, etc.).
Challenges arising from alternative investment management
20 Banque de FranceFinancial Stability Review No. 3 November 2003
Ackermann (C.), McEnally (R.) and Ravenscraft (D.)
(1999): “The Performance of Hedge Funds: Risk,
Return and Incentives”, Journal of Finance, vol. 54,
n°3, p. 833-874
Agarwal, (V.) (2000): “Generalized Style Analysis of
Hedge-funds”, The Journal of Asset Management,
vol. 1, n°1, p. 93-109, July
Agarwal (V.) and Naik (N. Y.) (2000a): “On taking the
Alternative’ Route: Risks, Rewards Style and
Performance Persistence of Hedge Funds”, Journal
of Alternative Investments, 2, p. 6-23, Spring
Agarwal (V.) and Naik (N. Y.) (2000b): “Multi-Period
Performance Persistence Analysis of Hedge Funds”,
Journal of Financial and Quantitative Analysis, 35,
p. 327-342
Agarwal (V.) and Naik (N. Y.) (2003): “Risks and Portfolio
Decisions Involving Hedge Funds”, Working paper
Alternative Investment Management Association
(2002): Guide to Sound Practices for European Hedge
Fund Managers, August
Amenc (N.) and Martellini (L.) (2003): “The Brave
New World of Hedge Fund Indices”, Working paper,
EDHEC, Misys Risk and Asset Management
Research Centre
Amenc (N.), Curtis (S.) and Martellini (L.) (2003a):
“The Alpha and Omega of Hedge Fund Performance
Measurement”, Working paper, EDHEC, Misys Risk
and Asset Management Research Centre
Amenc (N.), Martellini (L.) and Vaissié (M.) (2003b):
“Benefits and Risks of Alternative Investment
Strategies”, Journal of Asset Management, vol. 4, n°2,
p. 96-118
Amin (G.) and Kat (H.) (2003): “Hedge Fund
Performance 1990-2000: Do the Money Machines
Really add Value?”, Journal of Financial and
Quantitative Analysis, vol. 38, n°2, June
Asness (C.), Krail (R.) and Liew (J.) (2000): “Do
Hedge Funds Hedge?”, Working paper, AQR Capital
Association française de la gestion financière
(AFG-ASFFI) (2002): “Gestion alternative : recueil
Basel Committee on Banking Supervision (BCBS)
(2000): “Banks’ interactions with Highly Leveraged
Institutions: Implementation of the Basel
Committee’s sound practices paper”, January
Brealey (R.) and Kaplanis (E.) (2001): “Hedge Funds
and Financial Stability: An Analysis of their Factor
Exposures”, International Finance 4:2,p. 161-187
Brooks (C.) and Kat (H.) (2002): “The Statistical
Properties of Hedge Fund Returns and Their
Implications for Investors”, The Journal of Alternative
Investments, vol. 5, n°2, p. 26-44, Fall
Brown (S.) and Goetzmann (W.) (1997): “Mutual Fund
Styles”, Journal of Financial Economics, 43,
p. 373-399
Brown (S.), Goetzmann (W.) and Ibbotson (R. G.)
(1999): “Offshore Hedge Funds: Survival and
Performance 1989-1995”, Journal of Business, 72,
p. 91-117
Carhart (M. M.) (1997): “On Persistence in Mutual
Fund Performance” , Journal of Finance, vol. 52, n°1,
p. 57-82, March
Committee on the Global Financial System (CGFS)
(2003): “Incentive Structures in Institutional Asset
Management and their Implications for Financial
Markets”, March
Donaldson (W.) (2003): “Testimony before the Senate
Committee on banking, housing and urban affairs
concerning Investor protection implications of
Hedge Funds”, Document SS1, Securities and
Exchange Commission (SEC), April
Dybvig (P.) (1988a): “Distributional Analysis of
Portfolio Choice”, Journal of Business, vol. 61,
p. 369-395
Dybvig (P.) (1988b): “Inefficient Dynamic Portfolio
Strategies or How to Throw Away a Million Dollars
in the Stock Market”, Review of Financial Studies,
vol. 1, p. 67-88
Challenges arising from alternative investment management
Banque de FranceFinancial Stability Review No. 3 November 2003 21
Edwards (F.) and Caglayan (M.) (2001): “Hedge fund
and Commodity Fund Investment Styles in Bull and
Bear Markets”, Journal of Portfolio Management,
vol. 27, n°4, p. 97-108
Eichengreen (B.), Chadha (B.), Jansen (A.), Kodes
(L.) et Sharma (S.) (1998): “Hedge Funds and
Financial Market Dynamics”, Occasional Paper
n° 166, IMF
Eichengreen (B.) and Mathieson (D.) (1999): “Hedge
Funds: what do we really know?”, IMF Economic
Issues, n° 19
Fama (E. F.) and French (K. R.) (1996): “Multifactor
Explanations of Asset Pricing Anomalies”, Journal of
Finance, vol. 51, n°1, p. 55-84, March
Ferson (W. E .) and Schadt (R. W.) (1996): “Measuring
Fund Strategy and Performance in Changing
Economic Conditions”, Journal of Finance, vol. 51,
n°2, June
Financial Services Authority (FSA) (2003): Hedge
Funds and the FSA, Discussion Paper 16 August 2002,
feedback statement on Discussion Paper 16, March
Fung (W.) and Hsieh (D. A.) (1997): “Empirical
Characteristics of Dynamic Trading Strategies: The
Case of Hedge Funds”, The Review of Financial
Studies, vol. 10, n°2, p. 275-302, Summer
Fung (W.) and Hsieh (D. A.) (2000): “Performance
Characteristics of Hedge Funds and Commodity
Funds: Natural versus Spurious Biases”, Journal of
Quantitative and Financial Analysis, 35, p. 291-307
Fung (W.) and Hsieh (D. A.) (2001): “The Risk in
Hedge Fund Strategies: Theory and Evidence from
Trend Followers”, Review of Financial Studies, vol. 14,
n°2, Summer
Fung (W.) and Hsieh (D. A.) (2002a): “Benchmarks
of Hedge Fund Performance: Information Content
and Measurement Biases”, Financial Analysts Journal,
vol. 58
Fung (W.) and Hsieh (D. A.) (2002b): “Asset-Based
Hedge-Fund Styles and Portfolio Diversification”,
Financial Analysts Journal, vol. 58, n°5,
Fung (W.) and Hsieh (D. A.) (2002c): “The Risk in
Fixed-Income Hedge Fund Styles”, Journal of Fixed
Income, 12, p. 6-27
Glosten (L.) and Jagannathan (R.) (1994):
A Contingent Claim Approach to Performance
Evaluation”, Journal of Empirical Finance, 1, p. 133-160
IOSCO (2003): “Regulatory and Investor Protection
Issues Arising from the Participation by Retail
Investors in (Funds of-) Hedge Funds”, Report of the
Technical Committee, February
Jensen (M. C.) (1968): “The Performance of Mutual
Funds in the Period 1945-1964”, Journal of Finance,
vol. 23, p. 389-419, May
Jorion (P.) (2000): “Risk Management Lessons from
Long Term Capital Management”, European
Financial Management6, September
Jorion (P.) (2001): “Value at Risk”, 2nd Edition,
Mc Graw Hill
Lhabitant (F. S.) (2001): “Assessing Market Risk For
Hedge Funds and Hedge Funds Portfolios”, Journal
of Risk Finance, p. 1-17, Spring
Lhabitant (F. S.) (2002): “Hedge Funds: Myths and
Limits”, Wiley Finance
Lhabitant (F. S.) (2003): “Hedge Funds: A Look
Beyond the Sample”, Working paper, EDHEC Risk
and Asset Management Research Centre
Liang (B.) (1999): “On the Performance of Hedge
Funds”, Financial Analysts Journal55, p. 72-85
Liang (B.) (2000): “Hedge Funds: the Living and the
Dead”, Journal of Financial and Quantitative Analysis,
September, p. 309-336
Liang (B.) (2001): “Hedge Fund Performance:
1990-1999”, Financial Analysts Journal, p. 11-18,
Liang (B.) (2003): “The Accuracy of Hedge Funds
Returns”, Journal of Portfolio Management,
Lo (A.) (2001): “Risk Management for Hedge Funds:
Introduction and Overview”, Financial Analysts
Journal, vol. 57, p. 16-33
Lobosco (A.) (1999): “Style Risk Adjusted
Performance”, Journal of Portfolio Management,
vol. 25, n°3, p. 65-68, Spring
Challenges arising from alternative investment management
22 Banque de FranceFinancial Stability Review No. 3 November 2003
Malkiel (B.) (1995): “Returns from Investing in
Equity Mutual Funds: 1971 to 1991”, Journal of
Finance, vol. 50, n°2, 1995, p. 549-572
Mitchell (M.) and Pulvino (T.) (2001):
“Characteristics of Risk and Return in Risk
Arbitrage”, Journal of Finance, vol. 56, n°6, December
Modigliani (F.) and Modigliani (L.) (1997): “Risk
Adjusted Performance”, Journal of Portfolio
Management, p. 45-54, Winter
Park (J.) (1995): “Managed Futures as an Investment
Set”, Doctoral Dissertation, Columbia University
Park (J.), Brown (S.) and Goetzmann (W.) (1999):
“Performance Benchmarks and Survivorship Bias for
Hedge Funds and Commodity Trading Advisors”,
Hedge Fund News, August
Schneeweis (T.) and Spurgin (R.) (1998): “Multifactor
Analysis of Hedge Funds, Managed Futures and
Mutual Fund Return and Risk Characteristics”,
Journal of Alternative Investments, 1, p. 1-24
Schneeweis (T.) and Spurgin (R.) (2000): “Quantitative
Analysis of Hedge Funds and Managed Futures
Return and Risk Characteristics”, Working paper
CISDM, University of Massachusetts, Amherst
Securities and Exchange Commission (SEC):
“Investor protection implications of hedge funds”
Sharpe (W. F.) (1966): “Mutual Fund Performance”,
Journal of Business, p. 119-138, January
Sharpe (W. F.) (1992): “Asset Allocation: Management
Style and Performance Measurement”, Journal of
Portfolio Management, vol. 18, p. 7-19, Winter
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Since hedge funds specify significant lock-up periods, we investigate persistence in the performance of hedge funds using a multi-period framework in which the likelihood of observing persistence by chance is lower than in the traditional two-period framework. Under the null hypothesis of no manager skill (no persistence), the theoretical distribution of observing wins or losses follows a binormial distribution. We test this hypothesis using the traditional two-period framework and compare the findings with the results obtained using our multi-period framework. We examine whether persistence is sensitive to the length of return measurement intervals by using quarterly, half-yearly and yearly returns. We find maximum persistence at the quarterly horizon indicating that presistence among hedge fund managers is short term in nature.
Although risk management has been a well-plowed field in financial modeling for more than two decades, traditional risk management tools such as mean-variance analysis, beta, and Value-at-Risk do not capture many of the risk exposures of hedge-fund investments. In this article, I review several unique aspects of risk management for hedge funds - survivorship bias, dynamic risk analytics, liquidity, and nonlinearities - and provide examples that illustrate their potential importance to hedge-fund managers and investors. I propose a research agenda for developing a new set of risk analytics specifically designed for hedge-fund investments, with the ultimate goal of creating risk-transparency without compromising the proprietary nature of hedge-fund investment strategies.
Using a large database, I studied hedge fund performance and risk during an almost 10-year period from 1990 to mid-1999. The empirical results show that hedge funds had an annual return of 14.2 percent in this period, compared with 18.8 percent for the S&P 500 Index. The S&P 500 is much more volatile, however, than hedge funds as a whole. Annual survivorship bias for hedge funds was 2.43 percent. I examined year 1998 in detail because hedge funds were heavily affected by the global financial market tumble in that year. For example, the highest volatility for hedge fund returns occurred in 1998, and more funds died and fewer were born in 1998 than in any other year of the period studied. Few funds changed their fee structures. In those that did, the fee changes were performance related; poor performers lowered their incentive fees.
It is widely agreed that asset allocation accounts for a large part of the variability in the return on a typical investor&apos;s portfolio. This is especially true if the overall portfolio is invested in multiple funds, each including a number of securities. Asset allocation is generally defined as the allocation of an investor&apos;s portfolio among a number of "major " asset classes. Clearly such a generalization cannot be made operational without defining such classes. Once a set of asset classes has been defined, it is important to determine the exposures of each component of an investor&apos;s overall portfolio to movements in their returns. Such information can be aggregated to determine the investor&apos;s overall effective asset mix. If it does not conform to the desired mix, appropriate alterations can then be made. Once a procedure for measuring exposures to variations in returns of major asset classes is in place, it is possible to determine how effectively individual fund managers have performed their functions and the extent (if any) to which value has been added through active management. Finally, the effectiveness of the investor&apos;s overall asset allocation can be compared with that of one or more benchmark asset mixes. An effective way to accomplish all these tasks is to use an asset class factor model. After describing the characteristics of such a model, we illustrate applications of a model with twelve asset classes to analyze the performance of a set of open-end mutual funds between 1985 and 1989. ASSET CLASS FACTOR MODELS Factor models are common in investment analysis. Equation (1) is a generic representation:
This paper attempts to shed light on the `black box' called hedge funds via the style analysis technique developed by Sharpe (1992). The conventional style analysis cannot be directly applied to hedge funds as it imposes two constraints: first, the style weights have to be non-negative, and secondly, they have to add up to 100 per cent. In addition, the conventional style analysis does not provide any information about the statistical significance of the style weights. In this paper, we conduct a generalised style analysis of various hedge fund strategies by relaxing the constraints of the conventional style analysis, and examine the significance of style weights, as in Lobosco and DiBartolomeo (1997). We find that the generalised style analysis approach is more robust for estimating the risk exposures of hedge funds that take short positions in various asset classes and typically hold a significant part of their portfolio in cash.Journal of Asset Management (2000) 1, 93-109; doi:10.1057/palgrave.jam.2240007
Using a new database of hedge funds, this paper provides a comprehensive analysis of the risk-return characteristics, risk exposures, style analysis and performance persistence of various hedge fund strategies. We conduct a mean-variance analysis to find that a combination of alternative investments and passive indexing provides significantly better risk-return tradeoff than passively investing in the different asset classes. Using a broad asset class factor model, we find that the hedge fund strategies outperform the benchmark by a range of 6% to 15% per year. We infer the significant risk exposures of different hedge fund strategies using generalized style analysis and find results consistent with their investment objectives. Finally, using parametric and non-parametric methods, we examine persistence in the performance of hedge fund managers. We find a reasonable degree of persistence which seems to be attributable more to the losers continuing to be losers instead of winners continuing to be winners.
In this paper I derive a risk-adjusted measure of portfolio performance (now known as Jensen's Alpha) that estimates how much a manager's forecasting ability contributes to the fund's returns. The measure is based on the theory of the pricing of capital assets by Sharpe (1964), Lintner (1965a) and Treynor (Undated). I apply the measure to estimate the predictive ability of 115 mutual fund managers in the period 1945-1964 - that is their ability to earn returns which are higher than those we would expect given the level of risk of each of the portfolios. The foundations of the model and the properties of the performance measure suggested here are discussed in Section II. The evidence on mutual fund performance indicates not only that these 115 mutual funds were on average not able to predict security prices well enough to outperform a buy-the-market-and-hold policy, but also that there is very little evidence that any individual fund was able to do significantly better than that which we expected from mere random chance. It is also important to note that these conclusions hold even when we measure the fund returns gross of management expenses (that is assume their bookkeeping, research, and other expenses except brokerage commissions were obtained free). Thus on average the funds apparently were not quite successful enough in their trading activities to recoup even their brokerage expenses.
This paper investigates the returns on British collectible postage stamps over the very long run, based on stamp catalogue prices. Between 1900 and 2008, we find an annualized return on stamps of 6.7% in nominal terms, which is equivalent to an average real return of 2.7% per annum. Prices have increased much faster in the second half of the 1960s, the late 1970s, and the current decade. However, we also record prolonged periods of real depreciation, for example in the 1980s. As a financial investment, stamps have outperformed bonds, but underperformed stocks. After unsmoothing the returns on stamps, we find that the volatility of stamp prices approaches that of equities. There is mixed evidence that stamps are a good hedge against inflation. Once the problem of non-synchronous trading is taken into account, stamp returns seem impacted by movements in the equity market.