Pwc Report En

Published on June 2016 | Categories: Documents | Downloads: 39 | Comments: 0 | Views: 370
of 187
Download PDF   Embed   Report

EU investment funds

Comments

Content


European Commission
DG Internal Market
Investment funds in the European Union:
Comparative analysis of use of investment powers,
investment outcomes and related risk features in both
UCITS and non-harmonised markets

European Commission
DG Internal Market
 2
1. EXECUTIVE SUMMARY
This report describes the methodology developed by PwC to consider the impact of wider
investment powers following the introduction of the UCITS III
1
Directives, together with
the results of our work, which is based on a representative sample of 530 funds split over
9 countries: France, Germany, Ireland, Italy, Luxembourg, Poland, Slovakia, Spain and the
United Kingdom. Our sample includes 380 UCITS and 150 non-harmonised funds.
The analysis addresses the main developments within the UCITS population over a five-
year period but also within the non-harmonised fund environment, in order to identify
whether the line between UCITS and non-harmonised funds is still clear, especially
concerning the risks associated with investments in the respective investment vehicles.
The introduction of wider investment powers may indeed result in additional risks for retail
investors. Our report therefore focuses on two main categories of risks:
 The market risk as measured by the volatility;
 The other risks (investment and operational, please refer to section 5.2) considered in
the light of the portfolio composition and changes thereto over the last five years.
In addition to the analysis of the risks mentioned above, the results of a detailed risk
management survey enable us to understand how such risks are identified and managed by
asset managers while a comparative regulatory analysis shows how such risks are
supervised by Regulators.
 Performance and risk analysis
For all funds in our sample, we calculated over the last five years:
- Average annual performance by category: equity, bonds, money market and mixed
funds for UCITS, hedge funds, funds of hedge funds, real estate and other non-
harmonised funds;
- Average annual volatility by category;
- Average Sharpe ratio by category;
- Value at risk by category.
At first glance, these ratios seem more favourable to non-harmonised funds. However,
ratios like volatility and historical VaR of a fund are only indicators of the market risk
associated with an investment; if market risk is a key risk for investors, there are also some
other risks that need to be considered in order to assess the global risk profile of an
investment. These other risks therefore also need to be taken into account when comparing
UCITS and non-harmonised funds, and are further analyzed in the section dedicated to
portfolio composition.
1
The term “UCITS III” refers to Directive 2001/107/EC, “the profession directive” and Directive
2001/108/EC, “the product directive”, jointly referred to as “the UCITS III Directives” (or Directive).
European Commission
DG Internal Market
 3
In terms of performance, in addition to this global picture, our analysis showed that all
asset classes (for both UCITS and non-harmonised funds) have experienced at least one
year of negative returns with the exception of real estate funds, whose returns were
consistently positive between 2002 and 2006.
More specifically, within the UCITS environment, when comparing performances of funds
launched before and after 2002 (i.e. launched under the older or the modified directive), we
have not identified any significant difference in behaviour between both categories.
When considering the impact of use of wider investment powers, the portfolio composition
analysis shows that the use of derivatives by UCITS funds increased and a small
proportion of funds even invests intensively in these products. This specific fund
population does not seem to significantly outperform other types of fund but on the other
hand it does not show a higher level of market risk compared to other funds which use this
type of product less.
When it comes to assessing market risk (measured by volatility) globally, non-harmonised
funds appear to be less volatile than equity UCITS, whereas real estate investments seem to
offer a good balance between risk and return. However, we must also bear in mind that if
the real estate market did particularly well during the period under review, this was not
always the case before 1998. Volatility of Funds of Hedge Funds was somewhat stable.
This said, in general, low volatility of non-harmonised funds may also be explained by the
frequency of NAV calculation (often monthly), which mechanically tends to lower
volatility.
The differences noted between both populations (UCITS and non-harmonized funds) in
terms of performance / risk ratio are representative of the market risk only. The analysis of
the portfolio composition can bring a more comprehensive view of the similarities and
differences in terms of risk profile and changes thereto between both populations.
 Changes in portfolio composition
PwC analysed the portfolio composition of funds and changes thereto since 2002 on the
basis of 380 financial statements of UCITS and 150 non-harmonized funds for the year
ended 2006. For UCITS funds, this analysis shows an effective use of the wider investment
powers introduced by the UCITS III directives: the proportion of target funds in this
population has increased significantly, and the use of derivatives also seems to be more
intensive both in terms of types and quantity. This may have an impact on the overall risk
profile of the UCITS product compared with former regulations.
Concerning the use of derivatives, the description of the UCITS population and its
developments would evidence a higher degree of sophistication than for non-harmonised
funds, both in terms of range of derivative types used and in terms of quantity.
Even if embedded leverage resulting from the use of derivatives does not seem to have
increased significantly over the period under review, leverage risk must still be further
analysed together with other sources of risk more specific to OTC derivatives:
counterparty, liquidity and valuation risks.
European Commission
DG Internal Market
 4
Compared with the non-harmonised fund population, it appears that a limited number of
investment types are still not present in the UCITS environment. Given the features of
these investments, whose use is restricted to the non-harmonized fund universe, some risks,
such as liquidity risks, remain higher or specific to non-harmonised funds.
Investing in both UCITS or non-harmonized funds implies exposure to specific risks which
are not limited to market risks. If investor protection and risk monitoring are applicable in
both environments, the risk assessment tools do not seem to be the same in all instances.
 Risk Management Survey
In this global context, our survey of fund managers aims to identify the current practice in
Europe in terms of risk management and take a closer look at how risks are measured and
at what risk management tools have been developed for that purpose. In addition, this was
the opportunity for asset managers to assess the changes in the risk profile of their
products, with a specific focus on newly introduced investment powers.
Even if the portfolio composition analysis showed an increased use of derivatives, it is still
assessed as moderate (47% of respondents) within UCITS funds and almost 10% of
respondents show no exposure at all. The results are even lower for UCITS III-only fund
managers, where the perceived use of derivatives is very moderate and the percentage of
no exposure reaches 33%. Surprisingly, even for non-harmonised investment funds, the use
of derivatives is quite limited.
Leverage risk
The derivatives exposure is generally motivated by hedging purposes (92%) and almost
half the respondents use derivatives for trading or arbitrage purposes. The use of
derivatives for leverage is very limited. In practice, the extensive use of derivatives
allowed by UCITS III gives the possibility for sophisticated funds to net positions that will
be captured by the Value at Risk.
Valuation risk
The valuation risk has not been reported as significant for complex instruments in
sophisticated UCITS. Valuation issues are often considered carefully before new
investments are accepted. As shown by the subprime crisis, valuation risk may however be
more significant with investments which are deemed to be standard. In addition, it appears
that the ability to independently price an instrument before trading it is not always the rule
for non-harmonised funds.
Liquidity risk
The liquidity analysis is developed and applied but not to a large extent. We also observe
that VaR calculation does not always include the impact of liquidity. This is even more the
case for non-harmonised funds.
European Commission
DG Internal Market
 5
Counterparty risk
Counterparty risk is logically measured by most asset managers. This is imposed by
UCITS III for OTC derivatives. Counterparty risk measurement tends to be harmonised
among asset managers even if it is not harmonised by regulators, as shown by the
comparative Regulatory Analysis. Asset Managers tend to adopt the most conservative
approach when performing measurements.
Other risk management issues
Most of our respondents (68%) have a formal risk management committee in place. All
Asset Managers promoting sophisticated UCITS have a Risk Management Committee. The
Risk Management Committee is actively involved in approving new product strategies
(81%) and in every decision related to the risk measurement methodology.
A new products committee has been established by the majority of respondents (69%). The
main reason for rejecting new products refers to the impossibility of evaluating the new
product (19%).
With reference to the risks arising from UCITS and non-UCITS funds, the respondents are
equally divided, almost half (47%) of the respondents believe that there is a difference
between risks arising from UCITS and those arising from non-harmonised funds. The
interviews indicate that this difference is linked to liquidity. Non-UCITS funds will often
be less liquid, as the redemption process is longer and sometimes complex.
The use of VaR is widespread, not only for sophisticated funds. It is an essential tool for
Risk Management units. Moreover, all the asset managers in our sample, managing only
non-UCITS, including Hedge Funds and Funds of Hedge Funds, measure the Value at
Risk. Even if VaR ratios are properly calculated, the impact of liquidity is not always
assessed.
The Sharpe ratio is used across the industry even if it is frequently criticised for not being
appropriate. It is also used extensively for non-UCITS. For Funds of Hedge Funds, the
maximum draw-down is always used, but with other, more traditional ratios.
The main impact of UCITS III for the next 12 months is expected to be on products (76%),
followed by IT, Risk Management and Processes (59% each).
On the products side, we observe in 2007 an increase in the creation of highly sophisticated
funds. This sophistication is linked to the issuance of specific regulations clarifying the
rules for so-called “sophisticated” funds but also to investments in risk management
procedures and systems allowing asset managers to apply the investment process set forth
in UCITS III.
UCITS III has contributed to dramatic changes in portfolio composition and risk
management processes but we believe that the impact on performance and the global risk
profile is neutral.
European Commission
DG Internal Market
 6
 Comparative regulatory analysis
With regard to UCITS, PwC identified some differences in the legal framework governing
funds, which can have an impact on the portfolio composition of the funds.
On the regulatory side, main differences are observed for the so-called “trash ratio”,
concerning transferable securities and money market instruments other than those referred
to in paragraph (1) of article 19 (1) of the Directive. In that context, France appears to be
the most flexible country and Luxembourg has adopted the same position as France, albeit
not to the same extent.
We also observe significant differences as to eligible counterparties and/or valuation rules
for OTC derivatives for which France and Italy are more demanding.
When we consider the use of derivatives, we can also note a lack of homogeneity in
Europe, especially in the context of:
 The definition of sophisticated funds and the distinction with non-sophisticated funds;
 The commitment approach for calculating the global exposure on derivatives for
non-sophisticated funds;
 The parameters imposed by regulators for calculating the Value at Risk;
 The potential limitation of leverage for sophisticated funds;
 And the calculation method of counterparty risks for OTC derivatives.
Those differences on the use of derivatives are significant enough to demonstrate that the
harmonization of supervision for UCITS is not yet complete.
However, this lack of harmonization does not lead to regulatory arbitrage, according to
respondents. The choice of a fund domicile is more linked to the availability of specific
operational capabilities and the reputation of such domicile, together with the acquaintance
with the vehicle and demand by sales forces.
 Conclusion and perspectives.
In conclusion, face-to-face interviews enabled us to get further details on the
implementation of UCITS III. In the interviewees' experience, the transposition to UCITS
III was a lengthy process. Asset Managers implemented the new Directives as soon as
possible to benefit from a wider range of eligible assets (mainly the creation of funds of
funds, cash and money market funds).
However, 90% of interviewees decided to wait for further regulatory clarifications before
creating sophisticated UCITS. It must be noted that most regulatory updates were issued in
2007 in the most active countries in terms of sophisticated funds like France and
Luxembourg.
Derivatives were used extensively before UCITS III in those countries, but not always in
the context of harmonised funds. Overall, interviewees believe that the use of OTC
derivatives has grown at a rate 10% per annum since the implementation of UCITS III and
also that such OTC derivatives allow for more flexibility in terms of product structuring.
European Commission
DG Internal Market
 7
The impact on Risk Management is more important, as developed in the risk management
survey. Strong Risk Management Units were already in place, but they had to adapt to
more complex products and cope with strong demand for financial innovation.
The final impact is the “substance issue” for management companies located in
Luxembourg and Ireland. Asset managers using UCITS III funds for cross-border
distribution purposes decided to set up their management companies very quickly, often
within 6 months after the transposition of UCITS Directives into domestic laws.
All interviewees recognized that a convergence is visible between alternative investments
and sophisticated UCITS funds. UCITS III allows the structuring of funds replicating
several alternative investment strategies with absolute returns. We also observe the interest
of Prime Broker, a service provider essential to Hedge Funds managers for sophisticated
UCITS.
It is also important to note that UCITS vehicles are used extensively for distribution to
institutional investors. Some pension funds and insurance companies do require the
supervision and investor protection provided by UCITS III. However, when a
UCITS product is sold to retail investors only, risks features are different in that they are
more suitable to investors having no risk appetite.
Finally, even if the regulation allows some UCITS products to implement alternative
investment strategies, asset managers usually consider the suitability of their funds for
retail investors as well as the risk management process they need before launching such
funds.
In our sample, 80% of the asset managers consider that leverage is not appropriate for retail
investors, and it can be sold to institutional investors having a good understanding of it.
In practice, leverage is very limited even if leverage opportunities resulting from UCITS III
are understood very well in France, Luxembourg and, to a lesser extent, in Germany and
the UK.
We should make a distinction, however, between leverage on the market (or β) and
leverage of some arbitrage opportunities (or the α). The leverage of the β is never
considered appropriate, except for some institutional investors. The leverage of the α may
be appropriate but with strong risk models and with appropriate disclosure to investors. But
even in that case, such leverage would be limited, as shown in the portfolio composition
analysis.
A last point mentioned by 60% of interviewees is that there is no demand for leverage by
investors.
All interviewees managing only UCITS are very satisfied with the flexibility provided
under UCITS III, but they would welcome additional powers in the following areas:
 Authorized investments in term loans;
 Authorized investments in hedge funds;
 Granting the European passport to feeder funds.
European Commission
DG Internal Market
 8
However, all interviewees would welcome more harmonization concerning the notification
process and tax issues.
A strong interest has been expressed for a specific Directive with more eligible assets for
funds sold to institutional investors. A European passport for Hedge Funds in that context
could be an asset.
Fund of Hedge Fund managers are also keen on a European passport on the basis of its
strong track record and with appropriate liquidity so that such funds can meet redemption
requests.
An important point concerning asset managers’ expectations is that no one among
interviewees is clearly interested in a future Directive based on risk management principles
rather than on investment restrictions and eligible assets.
They explained this point as follows:
 Even if risk principles are used increasingly, it is important to keep investment
restrictions used as guidelines by asset managers.
 Investment restrictions and eligible assets are objective criteria, easy to understand for
investors and often required by institutional investors.
 The concept of UCITS is linked to investor protection and it becomes a very strong
concept outside Europe. Investors in Asia and South America invest a lot in UCITS
products because of the quality of supervision, the excellent track record and investment
restrictions which are easy to understand by regulators in those countries. Removing
investment restrictions would be a threat to the UCITS concept in that context.
 Finally, 70% of interviewees believe that the market is not ready, this includes some
asset managers but also most investors and regulators.
European Commission
DG Internal Market
 9
GLOSSARY
- UCITS III: UCITS III is meant for the UCITS Directive, which has been modified by
the Directive 2001/107/EC, “the profession directive” and Directive 2001/108/EC, “the
product directive”, together “the UCITS III Directives”.
- Sophisticated UCITS: the concept of sophisticated UCITS is mentioned in the
recommendation issued by the European Commission on April 27, 2004 and is related
to UCITS applying specific risk management process for calculating the global
exposure on derivatives instead of applying a traditional commitment approach. Such
risk management process is based on a value at risk and stress testing calculations.
- Non-harmonised funds: funds not submitted to the UCITS directives and not benefiting
from the notification process for their pan-European distribution.
- Regulators in the sampled countries:
- France: Autorité des Marchés Financiers (AMF)
- Germany: Bundesanstalt für Finanzdienstleistungsaufsicht (BAFIN)
- Ireland: Irish Financial Services Regulatory Authority (IFSRA)
- Italy: Comissione Nazionale per la Societa e la Borsa (CONSOB) and Bank of Italy
- Luxembourg: Commission de Surveillance du Secteur Financier (CSSF)
- Poland: Polish Securities Exchange Commission
- Slovakia: National Bank of Slovakia
- Spain: Comisión nacional del Mercado de Valores (CNMV)
- United Kingdom: Financial Services Authority (FSA)
 10
1. EXECUTIVE SUMMARY 2
2. INTRODUCTION 13
3. SAMPLING METHODOLOGY AND RESULTS 15
3.1 Definition of the sample 15
3.1.1 Information sources 15
3.1.2 Definition of the criteria used 15
3.2 Resulting sample 17
3.2.1 UCITS sample 18
3.3.2 Non-harmonised fund sample 19
3.3 Collection of financial statements 22
4. PERFORMANCE / RISK ANALYSIS 26
4.1 Objectives 26
4.2 Quantitative Results 26
4.2.1 Population and sample statistics 26
4.2.2 Descriptive Statistics 28
4.2.3 Performance and Market Risk Analysis 29
4.2.4 Market Risk analysis 31
4.2.5 Risk-adjusted analysis 34
4.2.6 Sophisticated funds 35
4.2.7 UCITS Funds of Funds vs. Funds of Hedge Funds 36
4.3 Conclusion on performance and risk 37
5. PORTFOLIO COMPOSITION AND ASSOCIATED RISKS 39
5.1 Portfolio composition 39
5.1.1 Analysis of the UCITS population 39
5.1.2 Analysis of the non-harmonised fund population 57
5.2 Risks associated with changes to the portfolio composition 65
5.2.1 Changes to the UCITS environment 65
5.2.2 Non-harmonised universe specificities 69
5.2.3 Managers’ point of view on risk management and on the changes of risk
profiles 71
5.3 Innovation in the UCITS Fund industry 73
5.4 Conclusion on portfolio composition analysis 75
6. UCITS AND NON-UCITS: COMPARISON OF THE REGULATIONS
BETWEEN A SAMPLE OF EUROPEAN MEMBER STATES 77
6.1 UCITS 77
6.1.1 Risk -spreading ratios affecting the market risk 78
6.1.2 Distinction between sophisticated and non-sophisticated UCITS 78
6.1.3 Commitment approach for the use of derivatives 80
6.1.4 Risk management approach in terms of VaR implementation 81
6.1.5 Leverage in the case of sophisticated UCITS 83
6.1.6 Counterparty risk 84
6.1.7 Short sales of derivative instruments and cover requirements 85
 11
6.1.8 Valuation of OTC derivative instruments 87
6.1.9 Management companies and delegation of functions 88
6.2 Non-UCITS 90
6.2.1 Liquidity risk 90
6.2.2 Leverage 91
6.2.3 Assets valuation 93
6.3 Conclusion 93
7. GENERAL CONCLUSION OF THE PROJECT 94
 12
APPENDICES
 APPENDIX A (p.98): Sampling methodology
 APPENDIX B (p.105): Methodology of performance/risk analysis
 APPENDIX C (p.117): Detailed geographical asset allocation of UCITS equity
funds
 APPENDIX D (p.120): Detailed of UCITS Bond and Money Market Funds
investments composition by currency.
 APPENDIX E (p.123): CSSF statistical reporting
 APPENDIX F (p.126): Importance of operational risk
 APPENDIX G (p.127): Survey on Fund failure
 APPENDIX H (p.129): Risk Management Survey
 APPENDIX I (p.160): Comparative regulatory analysis – additional information
 APPENDIX J (p.162): Alternative Investment Survey
European Commission
DG Internal Market
 13
2. Introduction
According to EFAMA, the European investment management industry has experienced
strong growth in assets under management with total EU assets of over EUR 7.6 trillion at
the end of 2006.
The UCITS directives of 1985 and 2001 have been the pillars of the European regulatory
framework and have contributed to the continuous success of UCITS both inside and
outside of Europe. The UCITS segment accounts for approximately 80% of the assets of
the investment fund industry in Europe and covers all funds under the UCITS III
Directives that implemented a “single license” regime. The remaining 20% of the market is
represented by non-harmonised funds, which are regulated according to national laws and
requirements. As from 1995, assets of non-harmonised funds grew by 15.1% annually,
reaching more than EUR 1,600 billion at the end of 2006.
Thanks to the recovery of the financial market since second semester 2003 and owing to
the implementation of the UCITS III Directives, which extended the range of eligible
assets, asset managers have been able to attract investors by boosting product innovation
and using new investment strategies.
With the development of new investment strategies and the increasing use of derivatives,
the border between UCITS and non-harmonised funds is becoming increasingly unclear. In
particular, many UCITS funds are now using alternative investment strategies through the
extensive use of derivatives and, to a certain extent, have come to resemble hedge funds.
In this context, where, in addition, competition is fierce and innovation a distinctive
competitive advantage, asset managers are increasingly demanding that their qualified
specialists identify new types of products that can be used as distinctive investment tools.
As a result of such sophistication, performance and related risks are top concerns for most
asset managers and focus on investment restrictions has become a must. Oxera (2006)
2
incidentally stressed the direct link which exists between the trend towards increased
product sophistication and diversity and the increase in the frequency and impact of
counterparty risk, misdealing, breach of client guidelines and fund valuation risk (please
refer to APPENDIX F (p.126) for additional details on this analysis)
Given this trend in the market and the changes in risk profiles, it is really interesting to
compare the outcomes (performance and related risks) of UCITS and non-harmonised
funds over the past years and identify the practical impact of such changes on the portfolio
composition of a large sample of EU funds. The section dedicated to changes in risk and
performance will focus on market risk while the second section will highlight other sources
of risk (e.g. investment and operational risks) through the analysis of portfolio
composition. It will then be interesting to consider whether asset management companies
have adapted their risk management framework and what requirements regulators impose
to address those risks.
2
Oxera, “Current trends in asset management”, 2006
European Commission
DG Internal Market
 14
The quantitative and qualitative approaches we have adopted are organized into four main
phases:
In Phase 1 we set out to analyze developments in market risk and performance for both
UCITS and non-harmonised funds by:
- Reporting on historical performance through a return analysis;
- Reporting market risk features for both UCITS and non-harmonised funds;
- Describing the potential impact of UCITS III implementation on performance and
market risk profile;
- Describing the potential differences in terms of performance and risk profile between
UCITS and non-harmonised funds.
In Phase 2 we set out to analyse portfolio composition by:
- Estimating the weight of transferable securities, money market instruments, derivatives
and non-financial assets into a similar sample of UCITS and non-harmonised funds;
- Classifying financial instruments;
- Identifying current portfolio composition trends and seeing what the current level of
sophistication (in terms of use of derivatives) is in UCITS and non-harmonised funds;
- Considering to which extent UCITS differ from non-harmonised funds in terms of asset
composition;
- Identifying sources of risks (in addition to market risk) associated with the differences
in asset portfolio composition.
In Phase 3 we will identify and assess the risk management procedures in place at
management companies or portfolio managers through:
- An online risk management questionnaire;
- Face-to-face interviews with asset managers of non-harmonised funds and UCITS
sophisticated funds.
In Phase 4 we will consider the regulatory environment of each country in the sample by:
- Reviewing the legal frameworks of the 9 countries and reporting the
specificities/additional requirements they have introduced;
- Identifying differences among the 9 countries in terms of eligible assets, use of
derivatives, and risk management requirements;
- Clarifying the regulatory environment for non-harmonised funds.
Before starting to present the analysis performed within each of the phases listed above,
together with their results, we will present our sampling methodology, together with the
resulting samples.
European Commission
DG Internal Market
 15
3. SAMPLING METHODOLOGY AND RESULTS
3.1 Definition of the sample
3.1.1 Information sources
Before starting the selection of the funds, a comprehensive view of the European
Investment Management market was helpful to identify the global population and its
components.
The data sources we used for this purpose are the following:
- EFAMA;
- LIPPER;
- The funds’ financial statements;
- Fitzrovia;
- Other applicable surveys and statistics.
For further information on these data sources and the specific use made of them in the
context of the survey, please refer to APPENDIX A (p.98).
Starting from this global picture, we needed to define a range of criteria in order to select
the funds to be included in our sample.
3.1.2 Definition of the criteria used
The sample that we were going to choose had to represent the European Investment
Management market as a whole. To ensure that the sample was representative, we looked
at several areas:
- The countries selected and their respective weight in the European Investment
Management industry;
- In terms of investment products, the sample needed to include both UCITS and
non-harmonised funds (including a variety of country-specific legal vehicles) and also
the range of investment strategies/vehicles available in the European Community.
Before further detailing our analysis, we would like to define the fund types that are
included under the respective UCITS and non-harmonised categories. In the context of our
analysis, we took the term UCITS to mean any fund which has a European passport and
which therefore complies with the requirements of the European Directives. A UCITS is
thus an harmonised fund, meaning that the products have similar features - whatever their
country of registration - and follow the same investment rules / restrictions (with some
local features, however, as described in section 6); they represent a homogenous European
population.
European Commission
DG Internal Market
 16
While there is one European definition of UCITS, the non-harmonised fund market
includes many product types, all of which may not be relevant in the context of the survey.
We considered the following funds within the scope of our analysis:
- Funds whose investment policy is not compliant with the investment rules defined by
the UCITS directives. In other words, funds whose investment policies are compliant
with UCITS requirements, but which do not have a European passport due to a
distribution network limited to the domiciliation country or outside Europe, were not
considered.
- Funds dedicated to institutional investors in accordance with the legal framework they
fall under, are also not part of the scope, as the survey aims to cover funds, which can
potentially include retail investors. Although this is a guiding principle, some
derogations to this rule have been accepted; these will be identified and justified when
applicable.
Within this European population of UCITS and non-harmonised funds, we selected the
countries to be surveyed on the basis of the following criteria:
- The local market needs to be developed both in terms of UCITS and non-harmonised
fund products, especially in countries that have joined the European Community since
its inception or some years ago (“the “older” European countries);
- As regards these “older” European countries, we selected the 7 countries that have the
most developed fund market (in terms of asset under management). Primary emphasis
was placed on the size of the UCITS market and the maturity of the non-harmonised
fund market was considered a secondary criterion. The assessment of the size of the
respective market has been based on the latest statistics provided by EFAMA, especially
concerning the description of the UCITS population;
- Concerning the new EU countries, which also needed to be included in the sample, we
selected the countries that have the largest and / or fastest-developing investment
management market;
- The weight of the respective countries in the selected population determined the number
of funds allocated to each country in our sample for both UCITS and non-harmonised
funds.
Based on these criteria, we were then able to start selecting both the UCITS and non-
harmonised funds that would be included in our sample. Even if the primary criteria were
similar for both types of funds, we had to use a different methodology in order to segment
the population because of the respective markets’ specific features.
The detailed statistical methodologies applied for the selection of UCITS and non-UCITS
funds within their respective categories are further detailed in APPENDIX A (p.98).
European Commission
DG Internal Market
 17
3.2 Resulting sample
EFAMA was used to identify the most active European markets as well as the growing
New Member States in the field of investment management.
The following countries were selected:
Domiciliation country Total assets of nationally domiciled
UCITS and non-UCITS
(in billion of EUR)
European market share
Luxembourg 1,525.2 23.2%
France 1,271.2 19.3%
Germany 967.2 14.7%
United Kingdom 644.5 9.8%
Ireland 584.5 8.9%
Italy 413.6 6.3%
Spain 275.1 4.2%
Poland 15.9 0.24%
Slovakia 3.3 0.05%
Total Europe 6,587.4 100%
Total selection 5,700.5 86.7%
Source: EFAMA 2006 Factbook, data as at 31/12/2005
Nine countries were included in our scope, two of which are New Member States. The
seven main markets are the largest in terms of assets in Europe and they account for more
than 86% of the total European market.
Among the New Member States, we decided to choose Poland and Slovakia. Why these
two countries? Poland is the most developed New Member State in the investment
management industry and the Slovakian fund market grew the most dramatically in Europe
between 2001 and 2005. The compound annual growth rate of assets reaches 100% during
this period and the ratio of total net sales in 2001-2005 totals 2,704%. Slovakia made
major gains in the IM industry and this progress justifies its inclusion within our analytical
sample.
Our sample was chosen based on the following picture of the European Investment
Management market, in terms of assets under management as at the end of
December 2005:
Source: EFAMA 2006 Factbook, data as at 31/12/2005
Within these nine markets, we defined a sample of funds both for UCITS and non-UCITS
items; the proportion of UCITS and non-harmonised funds was weighted according to the
market reality of each country in terms of number of UCITS and non-UCITS funds. As of
end of 2005, EFAMA recorded 44,498 investment funds, of which almost 80%
(31,181 items) were UCITS funds.
UCITS 79%
N-UCITS
21%
European Commission
DG Internal Market
 18
Based on the information and criteria described above, a first analysis shows that our
survey includes between 15% and 20% of the total European assets under management,
which represents a total sample of 400 to 500 UCITS and non-harmonised funds.
Given this, we set our minimum total sample to 500 funds, with an allocation between
UCITS and non-harmonised funds of minimum 350 and 150, respectively.
3.2.1 UCITS sample
The graphs below are based on EFAMA statistics as of December 2005 and concern the
selected countries of our sample. The weight of the UCITS population by country in terms
of number of funds is illustrated as follows:
Source: EFAMA
In terms of asset class invested in, the European fund market looks like this, according to
the number of UCITS funds:
Source: EFAMA
0.2%
0.4%
0.7%
4.0%
5.4%
8.5%
11.4%
29.7%
31.2%
8.4%
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Slovakia
Hungary
Poland
Italy
Germany
UK
Ireland
Spain
Luxembourg
France
Percent of funds
Percent of number of funds
6.5%
8.3%
21.3%
22.8%
41.0%
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45
Money market
Others
Bond
Balanced
Equity
Percent of funds
Percent of number of funds
European Commission
DG Internal Market
 19
The allocation between respective countries and asset classes is illustrated below, based on
the information contained in the 2 areas described above. Our sample covers a minimum of
350 UCITS funds.
UCITS France Germany Ireland Italy Lux. Poland Slovakia Spain UK Total
Equity 45 8 12 6 42 1 1 16 12 143
Bond 23 4 6 3 22 1 1 9 6 75
Balanced 24 4 7 3 23 1 1 9 7 79
Money market 7 1 2 1 7 1 1 3 2 25
Other 9 2 2 1 9 0 0 3 2 28
Total 108 19 29 14 103 4 4 40 29 350
The “Other” category mainly comprises guaranteed funds.
As indicated in our description of the UCITS population, we have struck a balance
between the largest funds and other smaller funds in every selected country, in order to
meet a minimum coverage of 20% of the total assets under management of the respective
domiciles.
3.3.2 Non-harmonised fund sample
Based on such primary criteria, we selected 184 non-harmonised funds (in order to ensure
we can report on a minimum of 150 funds), also taking into account:
- the domicile
- the legal structure
- the launch date
- the asset type (Lipper classification system)
- the fund size (asset under management)
We did not include in our sample any non-harmonised funds domiciled in Poland and
Slovakia for the following reasons:
- Poland: in terms of investment strategies, we understand that real estate funds exist
but are limited to closed-ended entities, which are not in the scope of our survey,
which focuses on open-ended structures as susceptible to be marketed to retail
investors. As far as funds of hedge funds are concerned, they were launched during
the year 2005 and do not have the minimum one-year track record required in the
context of our analysis; in addition, their number still appears to be very limited.
Finally, we did not find any evidence of private equity products in this market.
- Slovakia: there are really only a few real estate funds in Slovakia and these have
been launched very recently: out of the 4 real estate funds identified, 2 have been
launched in the last few months. Given the lack of history and the absence of a real
estate investment fund industry, we chose not to include this kind of fund in our
sample. In addition, we noted the existence of venture capital products, which have
been developed in order to support the transition of the Czech and Slovak Republics
to a global free-market economy. However, the shareholders of these funds are
institutional investors and they intend to support the development of these countries.
As a consequence, we did not consider these funds in our scope.
European Commission
DG Internal Market
 20
At this stage, such criteria are similar to those used to select the UCITS sample. The
distribution of non-harmonised funds in the selected countries, by type of investment
strategies and country is represented as follows, in terms of number of fund items:
Source: EFAMA
Source: EFAMA
Poland and Slovakia are not included in the above schedule due to the limited development
of the non-harmonised retail-oriented fund market. This decision is further justified above.
34.1%
39.0%
7.1%
4.4%
15.4%
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
FofHF
Hedge Fund
Other
Private Equity
Real Estate
Percent of funds
Percent of number of funds
23.6%
12.6%
16.5%
12.6%
19.8%
3.3%
11.0%
0.0 0.1 0.1 0.2 0.2 0.3 0.3
France
Germany
Ireland
Italy
Luxembourg
Spain
UK
Percent of funds
Percent of number of funds
European Commission
DG Internal Market
 21
Based on these 2 aspects, our total non-harmonised fund sample comprising 184 funds has
been allocated as follows between the respective countries and asset classes:
Non-
UCITS
France Germany Ireland Italy Luxembourg Spain UK Total
FoHF 13 6 9 14 19 1 62
Hedge
Fund
25 7 19 1 16 3 71
Other 5 7 12
Private
Equity
3 8 11
Real
Estate
10 2 1 1 6 8 28
Total 46 23 30 23 36 6 20 184
The following country-specific features have been taken into account when defining our
sample and have been supplemented by our analysis of the legal framework in section 6
below:
* France
The sample covers the following vehicles: ARIA, ARIA EL, Aria Alternatif (classified as
hedge funds), “Fonds à Formule” (classified under the “Other” category). France has also
developed a range of private equity-like products for retail distribution purposes (FCPR,
FCPI). Real estate vehicles do not appear to be types of funds, which have been broadly
developed for retail distribution purposes in France, so far.
* Germany
Our sample can almost be divided equally between alternative investment funds and
property-based vehicles through the strong presence of real estate funds among retail
investors.
* Ireland
The focus placed by private equity funds on institutional or qualified investors as well as
the typical closed-end legal form suggests that this form of investment vehicle should not
be considered in our analysis.
* Italy
The sample has been built considering the developed alternative investment fund
environment, even if it is not open for distribution to retail investors as such. The number
of real estate funds though, appears to be limited.
Funds classified under the “Other” category correspond to “Fondi Non Armonizzati
Aperti”.
* Luxembourg
In the Luxembourg market, the focus placed by private equity funds on institutional or
qualified investors suggests that this type of funds should not be considered in the analysis.
Hedge Funds and Funds of Hedge Funds available for more retail investors have been
considered however.
European Commission
DG Internal Market
 22
* Spain
The sample only contains real estate funds. While alternative investment funds have
developed recently in Spain, they were not relevant to our sample, since this type of
investment vehicle does not have a minimum track record of one year.
* United Kingdom
The sample includes a significant number of private equity and venture capital trusts, as
well as real estate products. The limited development of Hedge Funds and Funds of Hedge
Funds in that country seems to be justified by the absence of a specific legal framework,
which makes local promoters reluctant to launch such funds in their home country and opt
instead for an off-shore solution.
The criteria used to select the funds to be included in our sample and their corresponding
results have been described. We can now move on to the analysis of changes in the
NAV/share of the funds selected between 2002 and 2006, followed by a description of the
changes in the portfolio composition over the same period of time. These topics will be
addressed in sections 4 and 5, respectively.
Whereas all NAV information needed to perform a detailed analysis has been retrieved
from Lipper, we needed to obtain the financial statements of a sample of sub-funds, which
had the profile described in the above section. The following section describes the process
used to collect the financial statements as well as the results achieved in that context.
3.3 Collection of financial statements
As mentioned above, the objective is to survey a minimum sample of 350 UCITS and
150 non-harmonised funds domiciled in the 9 European countries selected.
As defined in our sampling methodology, we also decided to select funds with a launch
date before and after the introduction of the UCITS III directives. As a result, some of the
UCITS and non-harmonised funds may not have financial statements over a five-year
period. Since we selected funds which had a minimum track record of one year,
all selected funds have some financial statements produced in 2006.
The results presented in section 5 refer to the detailed analysis of 380 UCITS and
150 non-harmonised funds whose financial statements were actually collected in the 2006
financial year.
Except in Luxembourg and Spain, we have not identified any centralized national websites
where the financial statements of the investment funds domiciled in that country could be
easily retrieved. Therefore, in order to get the information, we had to contact the promoters
directly in their respective home country or consult their websites. Although the project
was welcomed by most of the promoters we contacted, in some instances, we did not
receive an answer to our query and in some others, we received a negative reply.
European Commission
DG Internal Market
 23
Five promoters of UCITS funds were not interested in the project and did not provide us
with the requested financial statements, for reasons that were never made clear. In the case
of non-harmonised funds, 15 promoters would not give us access to the financial
statements of their funds, arguing that the information was privileged and that the
documents could only be circulated to the funds’ shareholders. Although we emphasised
that the information made available to us as part of the survey was treated in strictest
confidence, these promoters would not budge. In those cases where we had to rely on the
information available on the promoters’ website, the oldest historical information (2002 to
2004) was not always accessible.
Once the financial statements were collected, our target was to capture the portfolio
composition of the respective funds in such a manner that would allow it to be easily
retrieved. The methodology and classification used to collect the information were the
same for UCITS and non-harmonised funds. As a consequence and in order to allow for a
meaningful comparison, we needed to find the right balance between the required level of
detail and the need to not be overly specific.
Thus, in agreement with the European Commission, the following key categories of net
assets / portfolio composition and relevant information to be captured were used:
- Equities: investment in equities was segregated into listed and unlisted investments. We
also had to distinguish between the domiciliation countries of the respective
investments. In order to avoid spreading investment portfolio over a too large number of
countries, and, given the likely inconsistent level of details contained in the financial
statements, we captured this information under the following categories: Americas,
Asia, Emerging Markets and, for Europe, we separately identified the 9 countries
included in our sample and other European countries. We included more detailed
information concerning European countries in order to capture to which extent, within
each of the selected domiciliation countries, portfolio managers invested in their home
market or in other European countries.
- Bonds and money market instruments: the domiciliation country of the issuer was not
considered the most relevant criterion in the context of our analysis, as the return on this
type of investment is primarily linked to the currency of the issue and not to the
nationality of the issuer. We were aware that this may have introduced a bias, especially
concerning issues made in USD, which is the most common currency, including for
high yield bonds and emerging country issues, which have a different risk profile and
return than “common” US bonds. As a consequence, the bond and money market
investments have been captured within 2 categories: high yield investments (including
funds with a high yield branding and emerging markets issues) and other bonds issued
in currencies of developed countries and with no specific risk profile identified.
In addition, the classification of bonds and money market instruments, as traded on a
regulated market or not, may differ as a result of different interpretations by national
supervisory authorities: the markets of exchange for this type of investment, even if
they are very active, are clearly not official markets as they are for equity investments.
The classification (listed / unlisted) as detailed in the financial statements has been used.
European Commission
DG Internal Market
 24
- Investments made in other funds (“Target funds”): the nationality of the management
company of target funds is not disclosed in the financial statements and has therefore
not been included in our analysis. In order to avoid spreading the surveyed population
over too many categories of investment strategies, we used the following profiles:
- UCITS: Bonds / Equities / Money Market / Mixed / High Yield - Emerging Markets.
- Hedge Funds: Long /short, Arbitrage and Relative Value, Event Driven, Directional,
Fund of Funds and Multi strategy.
- Other non-UCITS funds were classified based on the same categories as UCITS
funds.
The classification within the respective categories has been made based on the name of the
target funds.
- Derivatives: The purpose of the analysis is to identify changes in the use of derivatives
over the past 5 years. These changes need to be considered in terms of derivatives
invested in (are there some new types of derivatives?), in terms of the number of
derivatives used but also of number of lines in the portfolio, and finally in terms of
embedded leverage. We collected the information concerning derivatives in 2 stages:
- Input of accounting information (market value / unrealized gain or loss or some
derivatives) to obtain a global overview of portfolio composition from 2002 to 2006
and the changes in weight of the respective components over the same period of
time.
- For Funds having derivatives, we will capture the embedded leverage resulting from
the use of such derivatives over the same period of time. The level of embedded
leverage will take into account all the details, which are disclosed in the financial
statements.
The derivatives have been segregated according to the following criteria: listing location,
exposure, etc.
- Cash: the level of cash is also an important element of the net asset and performance of
the fund. Within this category, we included all components of liquid items (cash,
margin accounts) but also overdrafts and loans.
- Other elements that could not be included in the above-mentioned categories have been
inserted under the “other net assets” category. This section appears to be a minor
portion of total net assets, as reported in the graphs enclosed in section 5.
The level of detail of information that could be collected was highly dependent on the
disclosures made in the financial statements. We noted significant differences in the level
of detail available in the financial statements, depending on the country where the funds
are domiciled. In some instances, only a limited level of information could be retrieved
from the financial statements due to the lack of information or transparency of the
documents. This is particularly true of 2 countries:
- France where the disclosure of derivatives positions was not mandatory before 2005. In
some instances, we also relied on a semi-annual “bulletin”, which does not give all
detailed statistics on the investment portfolio, but only broad statistics.
- Spain where derivatives are not market valued but kept at cost in the Fund’s records. In
addition, in the notes to the financial statements, there is only limited information on the
nature and underlying of the derivatives used by the portfolio manager.
European Commission
DG Internal Market
 25
Based on this proposed methodology, the results in terms of portfolio composition in 2006
and changes in previous years are disclosed in section 5.
Before detailing the results, we will first analyse the performance and volatility resulting
from the analysis of historical NAV per share data.
European Commission
DG Internal Market
 26
4. Performance / Risk Analysis
4.1 Objectives
In this section, we introduce the methodology we have applied in order to evaluate
historical performances and we also present distinctive features in both risks and risk
adjusted performances of the selected funds. These elements will enable us to assess the
market risk associated with each type of vehicle / investment. In this section, the
reference to “risk” needs to be understood as “market risk” as other risks are further
analysed in section 5.2.
At fund level, traditional indicators are computed to facilitate direct and effective
comparison between all the different asset classes within the sample. These indicators are
further described in APPENDIX B (p.105). The study analyses the resulting indicators
(comparing results where applicable), and pays particular attention to:
- UCITS vs. non-harmonised funds;
- UCITS before 2002 vs. UCITS after 2002;
- Funds of UCITS funds vs. Funds of Hedge Funds;
- UCITS that use derivatives extensively (also called “sophisticated funds”) vs other
UCITS funds.
The sub-categories of the sample are also compared to ensure that the analysis is relevant
and that it takes into account the different types of assets under management (equities,
bonds, real estate, funds, etc).
The results of the analysis will provide more information about the historical
performances, market risk and risk adjusted performances of the European domiciled funds
over the five-year period under review (2002-2006) and on a yearly basis. Historical
performances are measured based on the calculation of the daily changes in NAV per
share.
The following section will provide with an analysis and discussion of the results obtained
from our sample of funds.
4.2 Quantitative Results
4.2.1 Population and sample statistics
The results of our performance analysis are based on the data obtained from a sample we
selected from the global Lipper population.
- Specific features of UCITS funds:
For UCITS funds, in order to verify the similarity between sample and population
distribution, we carried out a one-sample student’s t-test.
European Commission
DG Internal Market
 27
As a reference for the mean, we used the weighted statistics of the Lipper population
computed in accordance with the methodology described in APPENDIX B (p.105), so as
to be in line with the population described by EFAMA.
We set as a null hypothesis that population and sample means are equal, that is
Null Hypothesis H0: M=m.
Where
M is the population arithmetic mean
m is the sample arithmetic mean
For each asset type, we then computed a student’s t-test. If the p-value is less than 0.05, we
rejected the null hypothesis. The table below presents the results of our test.
Average Performance
(2002-2006)
One sample student’s t
test
Average Volatility
(2002-2006)
One sample student
test
Population Sample Student
t
N P-Value Population Sample Student
t
N P-
Value
Bond 2.85% 3.02% 0.70 89 0.49 3.87% 3.33% - 1.91 89 0.059
Equity 9.97% 10.47% 0.69 173 0.49 16.30% 16.93% 1.95 173 0.053
Mixed
Assets 6.34% 6.38% 0.10 97 0.92 6.96% 7.11% 0.38 97 0.705
Money
Market 1.31% 1.46% 0.33 34 0.75 2.32% 2.05% - 0.51 34 0.617
Other 4.28% 3.83% - 0.39 16 0.71 6.90% 5.50% - 1.66 16 0.116
As the test reported in the table fails to reject the null hypothesis for any of the statistics
indicated above, we cannot consider the differences between sample and population to be
statistically significant. We can therefore conclude that the statistics of the sample are
representative of the global population as available per LIPPER.
- Specific features of non-harmonized funds:
For this fund category, we could not test the representativeness of our sample against the
whole population, as we did for UCITS funds, since the categories we set (Hedge Funds,
Funds of Hedge Funds, Real Estate, Other) are not available in Lipper and the database
could not be reprocessed for this purpose.
European Commission
DG Internal Market
 28
For this population, however, we considered the potential impact of survivor bias. We
summarized this concept in APPENDIX B (p.105). To estimate this bias, we have analysed
the number of funds closed out for UCITS and non-harmonized funds over the 5 years of
the period under review, based on the whole population of funds available in the Lipper
database. The results are presented below: it appears that, even though a higher proportion
of non-harmonized funds was closed out during that period, we have not identified any
impact on the results that follow.
% of funds closed / total population available in Lipper
2002 2003 2004 2005 2006
Non-UCITS 8.96% 9.14% 9.73% 9.19% 9.36%
UCITS 3.34% 4.10% 3.54% 3.68% 3.37%
4.2.2 Descriptive Statistics
Table 1 reports summary statistics on each asset class over the whole period under review
(2002 – 2006).
TABLE 1: Basic Statistics of asset classes
This table summarises the average daily returns for the different asset classes from
January 1, 2002 through to December 31, 2006. The summary statistics include mean,
standard deviation (SD), median, skewness (Skew), kurtosis, minimum (Min) and
maximum (Max) of the daily annualized returns of our sample.
Mean (%) SD (%) Median (%) Skew Kurtosis Min (%) Max (%)
UCITS
BOND 0.013% 0.204% 0.020% -0.3223 6.5165 -1.022% 0.93%
EQUITY 0.037% 1.064% 0.075% -0.0771 4.0258 -5.256% 5.48%
MIXED ASSETS 0.023% 0.457% 0.044% -0.1816 5.2356 -2.327% 2.47%
MONEY MARKET 0.005% 0.127% 0.006% 0.8060 3.2932 -0.561% 0.64%
OTHER 0.014% 0.346% 0.022% -0.5534 8.9850 -1.760% 1.37%
Non UCITS
FUND of HF 0.028% 0.514% 0.030% -0.1777 5.8901 -2.61% 2.44%
HEDGE FUND 0.047% 0.702% 0.043% 0.0883 4.2845 -3.19% 3.30%
REAL ESTATE 0.133% 0.595% 0.035% 5.2564 45.1706 -2.53% 4.30%
OTHER 0.021% 0.426% 0.036% -0.4408 6.9056 -1.93% 2.03%
Over the period under review, the average daily return of UCITS funds is always below
those of European domiciled non-harmonized funds. However, if we look at Standard
Deviation (Volatility), the opposite is noted, confirming that non-harmonized funds may be
more volatile than UCITS. This is not true for equity funds however, for which volatility is
particularly high, due to the collapse of the Internet bubble 5 years ago.
On average, all UCITS asset classes, except money market, show a negative skewness
(asymmetric distribution), implying likelier but smaller gains. For hedge funds and real
estate funds, the combination of relatively high volatility and positive skewness would
suggest likelier but smaller losses and less likely but extreme gains. Within the UCITS
sample, bonds and “other” funds have significant fat tails. The value of kurtosis is rather
high, from 6.51 for bonds to 8.98 for “other”. When considering equity funds, we see that
neither skewness nor kurtosis are particularly high. The kurtosis, in particular, does not
significantly depart from normality (i.e. kurtosis equal to 3), which implies that the
geometric mean and standard deviation are representative of the features of the sample.
European Commission
DG Internal Market
 29
Interestingly however, the real estate funds also exhibit significant positive skewness,
suggesting that investing in real estate may offset downside risk.
4.2.3 Performance and Market Risk Analysis
Performance
In Table 2, we show the year-over-year results of different asset classes, thus illustrating
the trend followed by the performance during the whole period under review.
TABLE 2: Average performance of asset classes
Even if our survey focuses on the 2002-2006 period, we included the performance and
volatility information over a ten-year period, in order to have longer track records, and
include period where markets were not always in positive trends. This table presents the
mean returns on asset classes obtained from daily data for the 1997-2006 period converted
into annual numbers. The values represent a simple average of the annualised return of all
the funds included in each asset class category.
UCI TSAnnual Return(%)
Years 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Bond 7.75% 6.62% 5.52% 6.11% 4.75% 4.64% 2.62% 4.90% 4.16% 1.10%
Equity 26.69% 12.78%58.32% -3.56%
-
10.97% -24.49% 16.36%10.73%29.18%14.43%
MixedAssets 13.74% 9.79% 17.71% 4.16% -4.45% -9.79% 8.73% 6.43% 14.45% 6.79%
Money
Market
4.50% 4.12% 5.09% 5.04% 5.39% 0.68% -0.72% 0.85% 4.21% 1.45%
Other 8.97% 8.43% 10.67% 1.95% 1.79% -7.21% 1.10% 2.92% 8.19% 3.84%
Average
UCI TS 16.68% 9.76% 31.35% 1.25% -4.43% -13.06% 10.01% 7.42% 17.78% 8.35%
NonUCI TSAnnual Return(%)
Years 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
FoHF N.A. N.A. 62.93% 9.95% 9.16% -10.57% -2.48% -0.79% 16.99% 2.81%
Hedge
Fund -0.70% 15.1% 21.72% 7.88% 5.85% -5.35% 1.04% 1.59% 13.24% 2.22%
Real
Estate N.A. N.A. N.A. N.A. N.A. 5.22% 5.50% 10.47%10.86%14.18%
Other 36.59% 18.80% 15.45% -4.34% -3.17% -12.48% 6.31% 8.43% 13.31% 6.79%
Average
non
UCI TS 17.9% 15.9% 31.05% 6.63% 5.32% -7.70% 0.83% 2.44% 14.32% 4.71%
N.A.: no information is available due to lack of data for the years concerned. Our original
selection did not aim to cover a ten-year history but only the period from 2002 till 2006,
which explains the absence of data for some fund categories before 2002.
European Commission
DG Internal Market
 30
Because we cannot present individual results for each fund, we focused on the average
value of the funds included in each asset class. The annualised performances are measured
by an annualised value of the geometric mean.
Over the period under review, all asset classes experienced at least one year of negative
return with the exception of bonds and real estate funds, which consistently had positive
returns between 1997 and 2006; for the latter, no historical data is available before 2002.
In table 3, we illustrated the alphas for Equity and Bond asset classes obtained, using the
Capital Asset Pricing Model (CAPM).
TABLE 3: Alpha indicators for Equity and Bond asset classes (2002-2006)
This table features the alpha statistics on each asset class obtained from daily return data.
Alphas are estimated from the following regression, where R
BMK
is the daily return of the
most suitable market index based on the fund's investment style and geographical
allocation of the fund.
We identified the effect of systematic risk by performing a regression analysis for both
equity and bond funds, using the most appropriate market indices. The table displays the
alpha value to indicate the manager’s ability to obtain superior returns compared to the
market.
The figures represent average annual alphas obtained on the sample of funds. The average
R2 measures the soundness of our regression model. We do not show the results for the
other asset classes, as most of the alphas are not statistically significant (Low R2).
For equity and bond funds, alphas are relatively stable over the period under review.
However, R2 decreased regularly for funds launched after 2002. This may indicate a
departure from the behaviour of the benchmark. Interestingly, these funds experienced an
increased use of derivatives over the period. The use of derivatives to improve alphas is a
subject that will be discussed during face-to-face interviews.
Impact of UCITS III directives on performance
In table 4, we compare the performance and return figures for the UCITS funds launched
before and after 2002 after the introduction of the UCITS III directives. Money market
funds are not taken into account for the analysis, as the number of funds launched after
2002 is not sufficient to make statistically significant inferences.
t i t free BMK i i t free it
R R R R
, , ,
) ( c | o + ÷ + = ÷
European Commission
DG Internal Market
 31
TABLE 4: UCITS III comparative analysis (2002-2006)
The table presents annualised returns and volatilities on each asset class for funds launched
before and after the introduction of UCITS III.
Year
2004 56 11 4.95% 4.62% 3.39% 3.03% 152 10 10.70% 11.17% 13.15% 12.10%
2005 55 17 4.64% 2.61% 3.29% 2.03% 152 20 28.82% 31.95% 11.40% 12.24%
2006 56 20 1.22% 0.78% 3.05% 2.15% 152 22 14.38% 14.80% 14.02% 15.10%
Year
2004 65 10 6.65% 4.36% 5.89% 6.14% 12 14 4.56% 1.65% 5.74% 6.42%
2005 65 20 14.53% 14.12% 5.31% 6.17% 11 19 10.24% 6.96% 5.15% 5.47%
2006 66 22 7.08% 5.56% 6.47% 7.59% 11 22 5.11% 3.16% 4.74% 5.62%
Bond
Mixed Assets
Before After
Before After Before After Before After Before After
Equity
N Annualized return Annualised SD
Other
Before After Before After
Before After Before After Before After
N Annualized return Annualised SD
N
Before After Before After
Annualized return Annualised SD Annualized return Annualised SD N
Returns and volatilities are usually consistent for funds launched before and after the
implementation of the UCITS III directives. This would indicate that UCITS III had a
limited impact on performance and volatility. However, several factors may explain the
consistency in returns:
- UCITS III was not implemented simultaneously in all the countries in our sample;
- The detailed regulations governing the use of derivatives were often implemented
relatively late in some countries, as shown in the section on the comparative regulatory
analysis section;
- For existing funds, asset managers made the transition from UCITS I to UCITS III in a
number of different ways. How and how fast the transition to UCITS III was made will
be discussed in detail during face-to-face interviews.
4.2.4 Market Risk analysis
Tables 5 and 6 illustrate the different market risk measures calculated on the sample. In the
first table, we compared the traditional volatility measure. In Table 6, we present the
results obtained by the two VaR models mentioned: historical VaR and Parametric VaR,
thus showing the downside risk of each asset class. It should be stressed that all these
values represented an average of the funds within the sample and not an equally weighted
index of all funds of each asset class. Thus, for example, the 3.89% annual volatility of the
Bond class is the average of all funds’ volatility in the Bond asset class and not the
volatility on the average performance of the Bond asset class. The latter, corresponding to
a portfolio-like approach, would result in a very low volatility because of the large
diversification benefit. Since we cannot expect such diversification for individual
investors, it is unrealistic to show this low volatility as base volatility.
European Commission
DG Internal Market
 32
TABLE 5: Average annualised volatility (1997-2006)
This table presents the volatility on each asset class obtained from daily data for the years
over the period 1997- 2006 and converted into annual numbers. The annualised value is
calculated on a 260-day basis by applying the square-root-of-time rule: Annualised
volatility =σ*√260
UCI TSAnnual Volatility(%)
Years 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Bond 2.64% 2.98% 4.88% 4.92% 4.73% 3.89% 3.96% 3.34% 2.99% 2.81%
Equity 19.91% 23.77% 19.31% 22.48% 22.63% 25.24% 19.52% 13.09% 11.49% 14.15%
MixedAssets 7.69% 8.21% 8.47% 10.88% 9.07% 10.11% 8.92% 5.91% 5.47% 6.69%
MoneyMarket 0.54% 0.59% 1.83% 2.30% 2.55% 1.87% 2.13% 2.16% 2.14% 1.91%
Other 2.96% 2.50% 4.28% 5.94% 5.53% 6.21% 6.80% 6.12% 5.35% 5.32%
NonUCI TSAnnual Volatility(%)
Years 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
FoHF N.A. N.A. 13.39% 13.45% 10.44% 7.66% 7.81% 8.02% 7.61% 7.62%
HedgeFund 14.11% 19.1% 16.55% 17.41% 14.90% 13.68% 13.27% 11.20% 9.23% 9.05%
Real Estate N.A. N.A. N.A. N.A. N.A. 5.88% 4.58% 3.38% 4.66% 4.30%
Other 11.47% 12.06% 7.78% 9.17% 7.73% 12.51% 8.20% 5.86% 5.22% 6.45%
N.A.: no information is available due to a lack of data for the years concerned. Our
original selection did not aim to cover a ten-year history but only the period from 2002 till
2006, which explains the absence of data for some fund categories before 2002.
If we consider volatility as a good assessment of market risk, non-harmonized funds are
less volatile than equity UCITS, and real estate investments seem to provide a good
balance between market risk and return in that context. Volatility of Funds of Hedge Funds
was very stable.
European Commission
DG Internal Market
 33
TABLE 6: Value at Risk calculated at 1% significance level (2002-2006)
This table presents the results for the 1% significance level VaR analysis. We show both
the historical and parametric approach. Historical VaR is calculated through historical
simulation based on at least 100 past observations and taking the 1
st
percentile. Normal
VaR is computed following the RiskMetrics methodology, VaR=μ+z(α)σ considering z(α)
= -2.33 as volatility multiplier.
H-VaR
Normal
VaR
H-VaR
Normal
VaR
H-VaR
Normal
VaR
H-VaR
Normal
VaR
H-VaR
Normal
VaR
UCITS
BOND -0.60% -0.54% -0.67% -0.56% -0.53% -0.46% -0.45% -0.42% -0.43% -0.40%
EQUITY -3.90% -3.74% -2.91% -2.75% -2.24% -1.85% -1.97% -1.56% -2.58% -1.99%
MIXED ASSETS -1.56% -1.50% -1.38% -1.25% -1.05% -0.83% -0.92% -0.74% -1.24% -0.94%
MONEY MARKET -0.29% -0.27% -0.29% -0.31% -0.31% -0.31% -0.30% -0.29% -0.29% -0.27%
OTHER -1.08% -0.93% -0.90% -0.98% -0.94% -0.87% -0.82% -0.74% -0.90% -0.75%
Non UCITS
FUND of HF -1.35% -1.15% -1.06% -1.14% -1.24% -1.16% -1.19% -1.04% -1.36% -1.09%
HEDGE FUND -2.17% -2.00% -2.12% -1.91% -1.72% -1.61% -1.61% -1.28% -1.57% -1.30%
REAL ESTATE -0.83% -0.83% -0.54% -0.64% -0.43% -0.45% -0.54% -0.63% -0.66% -0.57%
OTHER -1.64% -1.59% -1.07% -1.03% -0.96% -0.81% -0.83% -0.71% -1.00% -0.81%
2002 2003 2004 2005 2006
As explained in the comparative regulatory analysis, the Value at Risk was implemented in
many countries for monitoring the global exposure of derivatives on sophisticated UCITS
funds. The use of Value at Risk followed a recommendation made by the European
Commission. However, no model or method was imposed. Therefore, it is interesting to
assess the Value at Risk for all the funds in the sample using these two different methods.
European Commission
DG Internal Market
 34
Based on the calculations made for our sample, we have the following comments:
- The results obtained using both methods are usually consistent, even if the differences
between the two methods are more visible for non-harmonized funds as shown in the
graph above;
- The results obtained using the historical method are often higher than the ones obtained
using the parametric method;
- Therefore, the historical method appears more conservative. It should be noted that the
results of this method are not affected by the potential non-normality of the distribution
of returns;
- Finally, even if the results are consistent between the two approaches, it appears that the
choice of a method impacts the results, which makes the supervision of sophisticated
UCITS more difficult in that respect.
4.2.5 Risk-adjusted analysis
Table 7 illustrates the traditional risk adjusted indicator developed by Sharpe. The Sharpe
Ratio does not refer to the market portfolio or any other benchmark. The implicit
benchmark is the risk-free rate of return.
TABLE 7: Sharpe Ratios of different asset classes (2002-2006)
This table presents the risk adjusted indicator based on the total risk (i.e. volatility) of the
funds. Sharpe ratios are estimated based on the return and volatility of each fund on an
annual basis. The risk free rate corresponds to the average six-month Euribor rate within
each year. As for the other figures shown in the performance and risk tables, the value
represents the average Sharpe ratio of all the funds in each asset class.
Risk in the context of the Sharpe Ratio is return volatility. The ratio measures the excess
return per unit of risk, thus the higher the Sharpe Ratio, the better the performance
generally. This is not the case when the excess returns are negative. In fact, this leads to
negative Sharpe Ratios that are difficult to interpret. This is the case for Money Market
funds because of the weight of expenses. But we do take into account expenses as they are
paid by investors. It is generally assumed that people prefer “more return” and “less risk”.
One would therefore expect that when ranking portfolios with equal returns by their Sharpe
Ratios, portfolios with lower volatilities are preferred to portfolios with higher volatilities.
European Commission
DG Internal Market
 35
The review of the Sharpe ratio for all asset classes under analysis shows that:
- Real Estate funds have a high risk adjusted performance profile over the period. As
mentioned above, we should however consider the positive cycle observed for that asset
class since 1998.
- The risk adjusted performance profile for hedge funds and funds of hedge funds is not
very high. The risk explained by the volatility is medium, but so are the returns.
- Equity and mixed UCITS assets are particularly interesting in terms of Sharpe ratio. But
Equity experienced a high level of volatility.
4.2.6 Sophisticated funds
Table 8 exhibits the results of the sub-sample of the so-called “sophisticated” funds based
on the definition we will present in the “Portfolio composition” (section 5.1).
The distinction between sophisticated and non-sophisticated UCITS is not always defined,
as explained in the comparative regulatory analysis. For the purpose of this study,
sophisticated funds shall be taken to mean any UCITS with more than 50 positions in
derivatives according to their portfolio composition as at December 31, 2006.
TABLE 8: Sophisticated Funds (2002-2006)
This table presents the performances, risks and risk adjusted indicators for the
sophisticated funds sub-category. Funds investing in fixed income securities (i.e. money
market and bonds on one side and equity and mixed assets on the other side) are merged.
Sharpe ratios are estimated based on the return and volatility of each fund on an annual
basis.
Years N
Annualize
d means
Annualize
d SD
Sharpe
ratio
HistVaR N
Annualize
d means
Annualize
d SD
Sharpe
ratio
HistVaR
2002 7 4.09% 3.90% 0.88 -0.59% 10 -21.19% 20.25% -1.20 -3.24%
2003 8 0.64% 3.86% 0.65 -0.58% 10 11.94% 14.95% 0.83 -2.15%
2004 8 3.33% 3.94% 0.71 -0.59% 11 6.16% 9.91% 0.52 -1.74%
2005 9 9.97% 4.54% 1.59 -0.62% 13 19.13% 7.77% 1.99 -1.26%
2006 9 0.29% 4.10% -0.48 -0.72% 13 7.52% 9.16% 0.45 -1.64%
Bond & Money Market Equity & Mixed asset
The first indication is that out of 380 UCITS funds in 2006, we have only 22 funds using
derivatives extensively (9 Bond and Money Market funds and 13 Equity and Mixed Assets
funds as detailed in table 8 above); These funds are among the largest funds in terms of
size in our sample. The portfolio analysis over the period under review needs to assess
whether derivatives were used to generate leverage. We can, however, observe that
volatility is stable (bonds and money market) or decreasing (equity and mixed assets).
It has to be noted that no Fund of Fund or funds belonging to the “other” category have
been identified within the sophisticated fund category.
In order to enable an effective comparison, we computed the same ratios for non-
sophisticated funds as for sophisticated funds with the same categories.
European Commission
DG Internal Market
 36
The last observation concerning “sophisticated funds” is that the historical VaR is low,
confirming the fact that the risk profile is not significantly different compared with non-
sophisticated funds.
For comparison purposes, we recomputed the ratios for non-sophisticated funds, based on
the same categories as the ones defined for sophisticated funds.
Based on this information, we can note that the general trend for both populations, i.e.
sophisticated and non-sophisticated funds, is similar, except in 2005 when bond and money
market funds outperformed non-sophisticated funds and 2005 and 2006 when
non-sophisticated equity and mixed asset funds more significantly outperformed
sophisticated funds.
Based on the limited size of the sophisticated sample, it is not possible to draw any
conclusion in terms of performance and risk of sophisticated funds versus
non-sophisticated funds. This analysis is only given as an indication based on data
available.
4.2.7 UCITS Funds of Funds vs. Funds of Hedge Funds
In table 9, we present a comparative analysis of funds of funds, namely UCITS funds of
funds vs. funds of hedge funds.
TABLE 9: UCITS funds of funds vs. funds of hedge funds
This table presents performance, risk and risk adjusted indicators for the fund of funds
sub-category. Sharpe ratios are estimated based on the return and volatility of each fund on
an annual basis.
Year
Annualised
return
Annualised
SD
Sharpe
Ratio
HistVaR
Annualised
return
Annualised
SD
Sharpe
Ratio
HistVaR
2002 -15.13% 14.67% -1.16 -2.37% -10.57% 7.66% -1.43 -1.35%
2003 13.66% 11.38% 1.15 -1.75% -2.48% 7.81% 0.01 -1.06%
2004 8.93% 8.36% 0.85 -1.36% -0.79% 8.02% 0.26 -1.24%
2005 17.27% 6.68% 1.91 -1.14% 16.99% 7.61% 1.84 -1.19%
2006 7.23% 8.13% 0.17 -1.50% 2.81% 7.62% 0.36 -1.36%
Funds of Funds Funds of Hedge Fund
N
Annualize
d means
Annualize
d SD
Sharpe
ratio
HistVaR N
Annualize
d means
Annualize
d SD
Sharpe
ratio
HistVaR
2002 79 3.28% 3.23% -0.27 -0.50% 201 -20.06% 20.76% -1.12 -3.20%
2003 84 1.57% 3.43% -1.06 -0.56% 212 14.26% 16.48% 0.84 -2.47%
2004 86 3.71% 2.96% -1.07 -0.46% 226 9.52% 10.87% 0.78 -1.87%
2005 93 3.60% 2.57% -2.30 -0.39% 244 24.57% 9.60% 2.26 -1.65%
2006 99 1.27% 2.40% -2.54 -0.36% 249 12.06% 11.77% 0.71 -2.15%
Bond & Money Market Equity & Mixed asset
European Commission
DG Internal Market
 37
We have the following comments regarding this table:
- The volatility of funds of hedge funds is stable over the period under review. For
UCITS funds of funds, this is the case as from 2003. However, we have to take into
account the fact that funds of hedge funds have often monthly NAV which
mechanically tends to lower the volatility;
- Except for 2002, the level of Value at Risk seems comparable between both categories;
- However, the Sharpe ratio is more favourable to UCITS funds of funds;
- Taking into account all the factors, funds of hedge funds would appear to be less risky,
but underperform UCITS funds of funds.
4.3 Conclusion on performance and risk
The basic statistics of the UCITS sample are representative of the global population, as
demonstrated by our statistical test.
The average performance calculations show that all asset classes (for UCITS and
non-harmonized funds) have experienced at least one year of negative return, with the
exception of real estate funds, whose returns were consistently positive between 2002 and
2006.
When comparing performances of funds launched before and after 2002, we cannot
identify a clear impact of UCITS III.
When it comes to assessing market risk, considering volatility an assessment of market
risk, non-harmonised funds are less volatile than equity UCITS. But as previously
explained for funds of hedge funds, most non-harmonised funds do not provide with a
daily NAV and when they have monthly NAV, their volatility is mechanically lowered.
Real estate investments seem to offer a good balance between market risk and return in
that context. However, we must bear in mind that the real estate market did particularly
well during the period under review, which was not always the case before 1998. Volatility
of funds of hedge funds was quite stable.
European Commission
DG Internal Market
 38
We have calculated the Value at Risk for all funds in our sample with two methods
(historical and parametric methods).
- The results are usually consistent, even if the differences between the two methods are
more visible for non-harmonized funds.
- However, the results obtained through the historical method are always higher than the
ones obtained through the parametric method.
- Therefore, the historical method appears more conservative. It appears that the method
selected impacts the results, which makes sophisticated UCITS more difficult to
supervise in that respect.
The graph above gives the positioning of all asset classes for UCITS and non-harmonized
funds in terms of market risk adjusted return and Value at Risk. It seems more favourable
to non-harmonized funds. However, the determination of the risk profile of a fund should
not be limited to the analysis of market risk based on historical performance. For that
purpose, some other sources of risk that might affect investors directly have to be taken
into account. The analysis of the portfolio composition may help us to identify new risks
introduced to the UCITS environment since the introduction of UCITS III. It may also help
us recognise risks specific to non-UCITS that may imply higher levels of risks than the
analysis of historical volatility of returns would suggest.
European Commission
DG Internal Market
 39
5. Portfolio composition and associated risks
5.1 Portfolio composition
Another aspect of the survey is to consider what impact the UCITS III directives had on
the portfolio composition of European investment funds and to which extent such funds are
similar to or different from non-harmonized funds. In order to answer this question, we
collected the financial statements of funds in accordance with the criteria and allocation
(by geographical area and asset type) defined above, from 2002 (or from their inception
date if they were launched after that date) to 2006.
First, our aim is to identify the main components of the Net Asset Value of the funds in the
surveyed countries and the changes thereto in both the UCITS and non-harmonized
environments. Second, we will consider changes to the risks associated with new
investments, first within the UCITS environment and then, through a comparison with
assets that are only present in the non-harmonized environment. This analysis will be
supplemented with some key results of the risk management survey and face to face
interviews we conducted in the context of this study. Finally, we will try to identify the
latest innovations in the UCITS environment, in terms of investment products, but also
investment strategies. This may evidence a potential convergence of the UCITS universe
with the “alternative” investment strategies, which were originally incompatible.
Within this section, all information will be disclosed in relative values, compared to the
population to which it refers.
5.1.1 Analysis of the UCITS population
The elements presented in these schedules result from the information disclosed in the
financial statements of the surveyed funds.
a) Global overview of Net Asset components.
Globally, UCITS funds classified by investment strategy were on average composed of the
following asset types, in the proportions indicated below:
NB: Other net assets correspond to all Net Asset components identified in the financial
statements and which do not belong to one of the main investment categories. They mainly
correspond to the net weight of accrued income and expenses.
European Commission
DG Internal Market
 40
UCITS FUNDS NET ASSETS’ COMPOSITION BY ASSET TYPE – 2002 - 2006
2002
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
Bonds Cash and cash equivalent Derivatives
Equities Funds Money Market Instruments
Other net assets Overdraft / loan Short / Sales
2004
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
Bonds Cash and cash equivalent Derivatives
Equities Funds Money Market Instruments
Other net assets Overdraft / loan Short / Sales
2006
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
Bonds Cash and cash equivalent Derivatives
Equities Funds Money Market Instruments
Other net assets Overdraft / loan Short / Sales
2003
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
Bonds Cash and cash equivalent Derivatives
Equities Funds Money Market Instruments
Other net assets Overdraft / loan Short / Sales
2005
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
Bonds Cash and cash equivalent Derivatives
Equities Funds Money Market Instruments
Other net assets Overdraft / loan Short / Sales
European Commission
DG Internal Market
 41
Based on this description, we can confirm that over the whole 2002-2006 period, funds
achieved their investment objectives by investing directly in asset types corresponding to
their investment strategy. Alternatively, funds were able to invest in other funds (“target
funds”) and in derivatives, to indirectly reach the same investment objectives. The use of
target funds corresponds to an alternative investment policy that is used in a proportion that
is not negligible, especially for mixed asset funds where it averages around 20% of the Net
Asset Value throughout the period. This highlights the fact that it is possible to have
significant investments in target funds under the UCITS III regulation, whereas this was
kept to a relatively low level under UCITS I. This is an important investment opportunity
that is used by most types of investment funds.
The importance of bonds within the Money Market category is linked to the use of
floating-rate bonds with quarterly or semi-annual reset of their interest rate, which implies
a closer adjustment to changes in interest rates.
The presence of short sales, which remain extremely limited, is noted in some French
UCITS only.
3
b) Geographical asset allocation
Given the reinforcement of the European Community, which is becoming a distinct and
effective single economic market, it is interesting to consider to which extent fund
managers remained focused, first, on their local equity markets in terms of investments and
second on the European Market.
Within the following schedules, the label “other” refers to other non-European countries or
portfolio components for which no geographical allocation could be defined based on
financial statement disclosures.
3
Please refer to section 6.1.1 for additional information on applicable legal requirements
European Commission
DG Internal Market
 42
EQUITIES – GEOGRAPHICAL FOCUS – 2002 -2006
2002
0%
20%
40%
60%
80%
100%
F
ra
n
c
e
G
e
r
m
a
n
y
I
r
e
la
n
d
I
ta
ly
L
u
x
e
m
b
o
u
r
g
P
o
la
n
d
S
lo
v
a
k
ia
S
p
a
in
U
K
Home Country European Countries Other
2004
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
la
n
d
I
ta
ly
L
u
x
e
m
b
o
u
r
g
P
o
la
n
d
S
lo
v
a
k
ia
S
p
a
in
U
K
Home Country European Countries Other
2006
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
re
la
n
d
I
ta
ly
L
u
x
e
m
b
o
u
r
g
P
o
la
n
d
S
lo
v
a
k
ia
S
p
a
in
U
K
Home Country European Countries Other
2003
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
Ir
e
la
n
d
I
ta
ly
L
u
x
e
m
b
o
u
r
g
P
o
la
n
d
S
lo
v
a
k
ia
S
p
a
in
U
K
Home Country European Countries Other
2005
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
Ir
e
la
n
d
Ita
ly
L
u
x
e
m
b
o
u
r
g
P
o
la
n
d
S
lo
v
a
k
ia
S
p
a
in
U
K
Home Country European Countries Other
European Commission
DG Internal Market
 43
APPENDIX C (p.117) includes a more detailed analysis of the countries present in the
equity portfolios of the surveyed funds.
From this description, we can identify the following major trends:
Over the selected period, the UK was the country, which primarily focused on its local
market, followed by France, Spain, Italy and Germany. Whereas the proportion of local
investment remained quite stable over the period in the latter countries, local investment
decreased from over 90% in 2002 to 75% in 2006 in the UK.
Luxembourg and Ireland have only limited investment in their local equities, which can be
explained by the limited size of their national equity market but also by the significant
weight of foreign or off-shore portfolio managers who may not specifically emphasise
investment on the fund’s country of domicile.
Given the limited size of the Polish and Slovakian sample (1 equity and 1 mixed asset
fund), we will not draw any conclusions from these statistics.
Concerning Bond and Money Market funds, they mainly invest in products issued in
currencies of developed countries; no specific developments have been identified in that
context. Additional information for the respective years is inserted in APPENDIX D
(p.120).
After this generic introduction on the changes in the components of the surveyed UCITS
funds, we will further investigate the changes in the use of derivatives, as the preliminary
analysis on global fund allocation did not enable us to determine a precise trend in that
context.
c) Use of derivatives
The UCITS modifying directives changed the limits applicable to the use of derivatives for
UCITS funds by moving from commitment restrictions to risk-oriented restrictions. Over
the same period, many new and innovative OTC derivatives were launched, which created
wider diversity and new opportunities for investment managers, who had been looking for
new investment strategies in the UCITS environment.
We will first analyze how the use of derivatives has developed in the various countries
over the surveyed period, in order to identify the main trends within the European
framework.
European Commission
DG Internal Market
 44
The following chart gives an indication of the changes to the weight of funds using
derivatives in our sample.
EVOLUTION OF FUNDS USING DERIVATIVES
Percentage of UCITS funds in our sample using derivatives (2002-2006)
0%
20%
40%
60%
80%
100%
2002 2003 2004 2005 2006
France
Germany
Ireland
Italy
Luxembourg
Spain
UK
No derivative products have been observed in the Slovakian and Polish funds included in
our sample.
In the other countries, this chart shows that an increasing proportion of funds within our
sample used derivatives between 2002 and 2006. Based on this first observation, we will
now analyse the trends, for the respective derivative types in order to identify what the
funds invest in, and consider if some specific observation can be made, depending on the
fund’s investment strategy or country of domicile.
European Commission
DG Internal Market
 45
- Presentation of the main derivative types and the changes to their weight
The following schedule provides information on how often the respective derivative types
can be found in the portfolio of the surveyed funds.
CHANGES TO THE PRESENCE OF DERIVATIVES
The figures in this table correspond to the percentage of UCITS funds (in number against
total population) in our sample using the following derivatives products from 2002 to
2006:
2002 2003 2004 2005 2006 Average
Futures 18% 24% 28% 34% 31% 27%
Repurchases/Reverse Repurchases 18% 15% 20% 22% 19% 19%
Options 14% 14% 18% 18% 20% 17%
FET 13% 15% 14% 18% 17% 15%
Swaps 1% 4% 4% 9% 12% 6%
Swaption 1% 1% 1% 1% 1% 1%
Contract For Difference (CFD) n.s. n.s. n.s. n.s. n.s. n.s.
“n.s.” stands for non-significant, as the number of observations for this category is too
limited and would result in a 0% disclosure, which would be misleading.
The analysis of this schedule confirms the increasing presence of derivatives in a larger
proportion of funds, especially futures, options and swaps. Repurchase / reverse repurchase
agreements have not increased dramatically. Futures, however, remain the most observed
derivative type and swap contracts grew the most over the period.
If the frequency of use is an interesting piece of information, statistics in terms of
accounting impact and weight in the fund’s net assets as presented in the following
schedule give another view.
European Commission
DG Internal Market
 46
The graph enclosed below lists the derivative types that were disclosed in the financial
statements of the surveyed funds together with their proportion within the derivative
population. The statistics are based on the weight of the respective derivative types, within
the fund’s net asset components linked to derivatives.
Type of derivatives – Average weight in derivative population
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2002 2003 2004 2005 2006
CFD FET Futures Options Repurchases/Reverse Repurchases Swap Swaption
The weight of the respective instruments was computed by taking into account the
unrealized gains or losses generated by CFDs (Contracts for Difference), FETs (Forward
Exchange contracts), Futures, Swaps and Swaptions. Concerning options, the market value
of the positions was considered. In the case of repurchases / reverse repurchase
agreements, the total amount corresponding to the value of securities to be settled was
used. These differences in treatment, especially concerning the latter derivative type, create
a significant gap in terms of weight in net assets of the respective derivatives. This
corresponds to an accounting reality and no conclusion relating to the exposure / risk
resulting from these differences should be drawn, as other derivatives imply a
commitment, which is not booked in the fund’s net assets and therefore not included in this
analysis.
In order to remove the bias, even more than repurchase / reverse repurchase agreements do
not give any additional market exposure, but provide a return in line with money market
instruments, we recomputed the same statistics, without taking this instrument into
account:
European Commission
DG Internal Market
 47
Type of derivatives – Average weight in derivative population
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2002 2003 2004 2005 2006
CFD FET Futures Options Swap Swaption
Based on this information, we can note that options remain the main derivatives in terms of
weight in net assets; throughout the period under consideration, they correspond to around
40% of the global net assets resulting from derivatives products. The increasing
importance of swaps is also evidenced in that context.
These preliminary results, based on unrealised gains or losses, or on market value for
options, may, however, conceal some specific developments by country or fund type that
will be further highlighted and analyzed in the following sections. The weight in net assets
may not be a key criterion to assess changes completely in the use of derivatives.
- Comparative analysis by country.
In order to obtain a more homogeneous and measurable population, we split the population
of the respective countries into a few categories: funds having no derivatives at all, those
investing in derivatives but in a limited proportion (less than 50 lines) compared to funds
disclosing up to 200 lines of derivatives in their portfolio. Another category was defined,
setting the limit over 200 lines. In the absence of a standard European definition of
sophisticated funds, in the context of our analysis, we considered sophisticated funds to be
any fund holding more than 50 lines of derivatives in its portfolio, based on financial
statements disclosures. In practice, it appears that funds classified as sophisticated based on
our internal criteria are also registered as sophisticated funds with their national regulator.
European Commission
DG Internal Market
 48
Based on this classification, the situation of the respective countries in the surveyed period
is the following:
Proportion of funds by level of sophistication
UCITS FUNDS – NUMBER OF LINES OF DERIVATIVE PRODUCTS – 2002-2006
2002
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
la
n
d
I
ta
ly
L
u
x
e
m
b
o
u
r
g
P
o
la
n
d
S
lo
v
a
k
ia
S
p
a
in
U
K
0 <50 50 to 200
2004
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
la
n
d
I
ta
ly
L
u
x
e
m
b
o
u
r
g
P
o
la
n
d
S
lo
v
a
k
ia
S
p
a
in
U
K
0 <50 50 to 200
2006
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
la
n
d
I
ta
ly
L
u
x
e
m
b
o
u
r
g
P
o
la
n
d
S
lo
v
a
k
ia
S
p
a
in
U
K
0 <50 50 to 200 over 200
2003
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
Ir
e
la
n
d
I
ta
ly
L
u
x
e
m
b
o
u
rg
P
o
la
n
d
S
lo
v
a
k
ia
S
p
a
in
U
K
0 <50 50 to 200
2005
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
la
n
d
I
ta
ly
L
u
x
e
m
b
o
u
r
g
P
o
la
n
d
S
lo
v
a
k
ia
S
p
a
in
U
K
0 <50 50 to 200 over 200
European Commission
DG Internal Market
 49
The information disclosed corresponds to the proportion of funds (in quantity) in the
respective categories, compared to the total population by country. Based on this chart, we
note that in all countries except the UK, more than 50% of funds hold derivatives in their
portfolio as from 2003, but in a proportion that remains limited, i.e. less than 50 lines. If
these charts evidence appearance of sophisticated funds as from 2002 in Ireland and
Luxembourg, 2005 in France and 2006 in Spain, Germany and Italy, their proportion
remains however very limited as less than 10% of the population. This said, the evolution
shows an increase in the proportion of these funds, together with introduction as from 2005
in France and 2006 in Luxembourg of funds using extremely intensively derivatives i.e.
over 200 lines in portfolio. Even if it does not correspond to a significant part of the
Investment Management European markets, this trends evidences an increasing interest for
this kind of products.
Whereas we noted that the frequency of use of all derivative types increased and that the
proportion of funds using derivatives followed the same trend, it is interesting to consider
into how many derivative types (independently from the number of lines in the portfolio)
funds invest on average. The situation is described in the table below.
Average number of derivative type held (2002-2006)
0
0.5
1
1.5
2
2.5
France Germany Ireland Italy Luxembourg Spain UK
2002
2003
2004
2005
2006
In that context as well, Ireland and Germany are countries where the use of derivatives is
the most important, as, as from 2005, funds domiciled there present more than 2 derivative
types in their portfolio in average, followed by Luxembourg.
European Commission
DG Internal Market
 50
The analysis of the population by country is a key criterion that enabled us to see the
differences in terms of evolution and level of sophistication within the respective countries.
However, this needs to be complemented by an analysis by fund investment type, in order
to consider whether the use of derivatives is dependent on the Fund type (i.e. equity, bond,
money market, mixed asset and other). In that context, we will not consider any specificity
in terms of country anymore, considering also that the international activities of most
promoters (who have funds domiciled in different European countries) make global
investment strategies similar, whatever the country of domiciliation of the funds. We are
aware that such promoters have to meet some specific requirements in terms of investment
restrictions, due to local specificities in the application of the Directives, which will be
further highlighted in section 6. We will however consider that this does not significantly
impact on the global population and therefore the results of our analysis.
European Commission
DG Internal Market
 51
- Comparative analysis by Fund asset type.
The following charts give an indication of the weight of the respective derivative types by
Fund asset type.
Types of derivatives by asset type – Average weight (%) in NAV – 2002-2006
2002
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
CFD FET Futures Options Repurchases/Reverse Repurchases Swap Swaption
2004
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
CFD FET Futures Options Repurchases/Reverse Repurchases Swap Swaption
2006
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
CFD FET Futures Options Repurchases/Reverse Repurchases Swap Swaption
2003
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
CFD FET Futures Options Repurchases/Reverse Repurchases Swap Swaption
2005
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
CFD FET Futures Options Repurchases/Reverse Repurchases Swap Swaption
European Commission
DG Internal Market
 52
In order to avoid the bias linked to the disclosure of repurchase / reverse repurchase
agreements as explained above, we made the same computations, without taking this
derivative type into account.
2002 - Without Repurchase / reverse Repurchase
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
CFD FET Futures Options Swap Swaption
2004 - Without Repurchase / Reverse Repurchase
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
CFD FET Futures Options Swap Swaption
2006 - Without Repurchase / Reverse Repurchase
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
CFD FET Futures Options Swap Swaption
2003 - Without Repurchase / Reverse Repurchase
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
CFD FET Futures Options Swap Swaption
2005 / Without Repurchase / Reverse Repurchase
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
CFD FET Futures Options Swap Swaption
European Commission
DG Internal Market
 53
The tables below present the weight of funds, by investment strategies, investing in
derivatives in the respective categories in 2002 and 2006.
2002
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
0 <50 50 to 200
2004
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
0 <50 50 to 200
2006
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
0 <50 50 to 200 over 200
Based on this information, we can note that “other” funds and bond funds were largely
using derivatives products: 87% and 77% of funds respectively in 2006. In terms of the
number of lines of derivatives present in portfolios, money market funds were the most
sophisticated in 2006. About 7% of funds in this category were even holding more than
200 lines of derivatives. When we consider the historical evolution of these proportions
since 2002, we could also stress that all categories of funds saw their proportion of funds
investing in derivatives significantly increase except equity funds where the proportion
remained stable over the past 5 years (45% on average).
2003
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
0 <50 50 to 200
2005
0%
20%
40%
60%
80%
100%
Bond Equity Mixed Assets Money Market Other
0 <50 50 to 200 over 200
European Commission
DG Internal Market
 54
Based on this broad overview, we will now consider in how many types of derivatives
funds included in our sample invest on average.
AVERAGE NUMBER OF DERIVATIVE TYPES HELD BY FUNDS – 2002-2006
0
0.5
1
1.5
2
2.5
Bond Equity Mixed Assets Money Market Other
2002
2003
2004
2005
2006
Bond funds as from 2005 and Money market funds as from 2006 reach the level of
2 derivative types present in average in fund portfolios. Within the “other” category, where
a large proportion of funds use derivatives, the average number of derivative types is
lower, even if it has increased since 2003; the limited use of derivative products by equity
funds is also confirmed in this framework.
In terms of frequency of presence of the respective main derivative types, by fund
investment strategy, the statistics are the following concerning the “most frequently used”
in 2006:
European Commission
DG Internal Market
 55
The figures below consider the percentage of UCITS funds of our sample that were using
the following derivatives products in for the respective years.
Proportion of funds within surveyed population investing in respective derivative
types from 2002 to 2006.
2002
FET Futures Options
Repurchases/Reverse
Repurchases
Swap Swaption CFD
Bond 30% 19% 15% 17% 2% 4% 0%
Equity 6% 17% 9% 21% 0% 0% 0%
Mixed Assets 16% 24% 20% 13% 0% 0% 0%
Money Market 5% 5% 11% 26% 0% 0% 0%
Other 0% 20% 40% 0% 40% 0% 0%
2003
FET Futures Options
Repurchases/Reverse
Repurchases
Swap Swaption CFD
Bond 33% 35% 18% 14% 7% 4% 0%
Equity 7% 17% 9% 16% 1% 0% 0%
Mixed Assets 17% 31% 23% 10% 4% 0% 2%
Money Market 5% 11% 11% 21% 0% 0% 0%
Other 0% 14% 29% 29% 43% 0% 0%
2004
FET Futures Options
Repurchases/Reverse
Repurchases
Swap Swaption CFD
Bond 24% 32% 27% 24% 4% 3% 0%
Equity 5% 20% 10% 15% 1% 0% 0%
Mixed Assets 22% 41% 26% 22% 3% 0% 0%
Money Market 10% 24% 10% 31% 10% 0% 0%
Other 9% 27% 27% 18% 36% 0% 0%
2005
FET Futures Options
Repurchases/Reverse
Repurchases
Swap Swaption CFD
Bond 33% 43% 30% 25% 17% 5% 0%
Equity 10% 25% 12% 20% 2% 0% 1%
Mixed Assets 20% 47% 23% 20% 6% 0% 3%
Money Market 10% 33% 10% 30% 23% 0% 0%
Other 15% 23% 23% 15% 23% 0% 0%
2006
FET Futures Options
Repurchases/Reverse
Repurchases
Swap Swaption CFD
Bond 34% 41% 27% 25% 23% 5% 0%
Equity 8% 23% 13% 15% 4% 0% 0%
Mixed Assets 20% 36% 30% 17% 9% 0% 0%
Money Market 11% 29% 14% 32% 25% 4% 0%
Other 13% 38% 19% 25% 25% 0% 6%
European Commission
DG Internal Market
 56
 Futures contracts are the most commonly used derivative type, except for money market
instruments where Repurchase / Reverse repurchase agreements are in first place.
 Repurchase / Reverse repurchase agreements are also widely used. Money market and
bond funds are the main categories of funds that use that kind of derivatives.
 Options are principally used by mixed assets and bond funds (between 25% and 30% of
these categories since 2004).
 Forward contracts are also types of derivatives mostly used by mixed assets and
especially bond funds with respectively 20% and 34% of funds in these categories using
forward contracts in 2006.
 Swaps are principally used by “other funds” and increasingly used by money market
and bond funds. Indeed 25% of money market funds and 23% of bond funds were using
swap contracts in 2006 compared to respectively 10% and 4% in 2004 and rather none
in 2002.
d) Evolution of leverage resulting from the use of derivatives.
The analysis of the portfolio composition evidences an increasing use of derivatives, in
terms of weight in net assets, but more importantly in terms of number of derivatives used.
In that context, it is interesting to analyse the evolution of the leverage resulting from the
use of derivatives, within the population using extensively derivatives (i.e. over 50 lines in
portfolio based upon our internal definition).
“Instrument leverage” is defined as the additional market exposure arising from off-
balance sheet exposure through derivatives products and was computed based on the
commitment resulting from open positions on derivatives. To estimate this type of
leverage, we computed a commitment ratio on derivatives products for sophisticated funds
i.e. funds with more than 50 lines of derivatives in 2006. Foreign exchange derivatives and
derivatives identified as used for hedging purposes in the financial statements were
excluded from our analysis. We remind that, in the context of our analysis, the term
sophisticated fund should not be understood as a legal concept (please refer to the
comparative regulatory analysis), but as the number of derivatives used as defined above.
European Commission
DG Internal Market
 57
Average commitment ratio on derivatives products for sophisticated
UCITS (2002-2006)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2002 2003 2004 2005 2006
As shown above, the average commitment ratio on derivative products of sophisticated
UCITS is moderate at 1.6x the NAV in 2006 in average. Between 2002 and 2006, this ratio
has been slowly growing since 2003 and ranges from 1.2x to 1.6x. This would indicate a
limited use of derivatives to leverage fund exposure to market.
Our analysis also showed an extensive use of derivatives as part of strategies rather than
additional exposure compared to portfolio composition (“naked position”), notably when
selling options. In practice, covered call/put options seem to be preferred to “naked”
call/put options and strategies that imply several options e.g. option spreads often result in
limited risk profile.
The Risk Management Survey part of our study will address in a more precise manner the
level of leverage and risk management developed by market participants who use
derivatives products.
5.1.2 Analysis of the non-harmonised fund population
The non-harmonized fund population was not impacted directly by the introduction of the
UCITS III directives, except some fund of funds and cash funds, which were non-
harmonized funds in the past. They did not meet the investment restriction under the
UCITS I regulation but they can now benefit from the UCITS passporting in the context of
the new Directives. The comparison of the portfolio composition of UCITS and non-
harmonized funds is however interesting in order to identify the type and weight of
investments present in the non-harmonized fund environment that remain not eligible
under the UCITS III directives, and to identify any trend in the way assets, that are eligible
for both types of funds, are used.
European Commission
DG Internal Market
 58
In that context, an analysis by country does not appear appropriate given the disparity in
terms of investment vehicles and investment strategies observed in the respective surveyed
countries. We will therefore focus our analysis on a range of criteria defined by investment
strategy:
- Hedge Funds;
- Funds of Hedge Funds (“FoHF”);
- Private Equity;
- Real Estate;
- Other.
Private Equity Funds and Real Estate Funds are investment vehicles that typically focus on
one type of investment on the market and look for investment opportunities within this
specific segment. In a different way, funds included within the Hedge Fund category can
potentially invest in a wider range of asset types, some of them being eligible for UCITS
funds, but are defined by their specific investment strategy or techniques. Funds of hedge
funds focus on a specific investment type (hedge funds), but may also seek additional
performance by using other investment instruments, such as derivatives, for instance.
European Commission
DG Internal Market
 59
Based on the financial statements, the main elements present in the non-harmonized funds
in the respective years, can be summarized as follows:
Net Asset Components – Non UCITS Funds – 2002-2006
2002
0%
20%
40%
60%
80%
100%
FoHF Hedge Fund Other Private Equity Real Estate
Bonds Cash and cash equivalent Commodities
Derivatives Equities Funds
Money Market Instruments Other net assets Overdraft / loan
Private Equity Real Estate Short / Sales
2004
0%
20%
40%
60%
80%
100%
FoHF Hedge Fund Other Private Equity Real Estate
Bonds Cash and cash equivalent Commodities
Derivatives Equities Funds
Money Market Instruments Other net assets Overdraft / loan
Private Equity Real Estate Short / Sales
2006
0%
20%
40%
60%
80%
100%
FoHF Hedge Fund Other Private Equity Real Estate
Bonds Cash and cash equivalent Commodities
Derivatives Equities Funds
Money Market Instruments Other net assets Overdraft / loan
Private Equity Real Estate Short / Sales
2003
0%
20%
40%
60%
80%
100%
FoHF Hedge Fund Other Private Equity Real Estate
Bonds Cash and cash equivalent Commodities
Derivatives Equities Funds
Money Market Instruments Other net assets Overdraft / loan
Private Equity Real Estate Short / Sales
2005
0%
20%
40%
60%
80%
100%
FoHF Hedge Fund Other Private Equity Real Estate
Bonds Cash and cash equivalent Commodities
Derivatives Equities Funds
Money Market Instruments Other net assets Overdraft / loan
Private Equity Real Estate Short / Sales
European Commission
DG Internal Market
 60
Within the above split of non-harmonized funds by investment strategy, we notice that:
Investments or techniques, which can only be used by non-harmonized funds in the
surveyed population, are:
- Short sales, which is an investment technique primarily used by Hedge Funds;
- Real Estate, which are only present in real estate funds;
- Commodities, which are only present in Hedge Funds;
- Private equity, which even if they correspond to equity investments, are, in most
instances, not listed and not liquid enough to be invested by UCITS funds.
The impact on the overall portfolio performance and associated market risk of these
investments or techniques were analysed in section 4. However, some other specific risks
may derive from these and will be analysed in section 5.2 below.
Other investments which are not specific to non-UCITS funds and can also potentially be
used by UCITS funds are the following:
- Equity and bond investments are not categories we will put a particular emphasis on:
these can be used by both UCITS and non-harmonized funds. The sole distinction is the
requirements in terms of listing and liquidity that is extremely important for UCITS
funds and can be derogated by non-harmonized. However the information is not clearly
disclosed in the financial statements of non-harmonized funds.
- Funds: the nature of target funds used by non-harmonized funds may, however, defer
from the fund types UCITS invest in. Indeed, UCITS III regulation defined key criteria
to be met so that investments in a fund can be authorized for UCITS funds. A more
detailed analysis of the specific risks associated with this kind an investment will also
be conducted in the following section.
- Cash and cash equivalent: if UCITS are now authorised to invest a significant portion of
their assets in time deposit and hold ancillary cash balance, they are not allowed to
borrow money; the active use of overdrafts / loans is only possible for non-harmonized
funds. Overdrafts are acceptable up to 10% of the Fund’s net assets for UCITS funds
but cannot be used as an investment facility.
Based on collected information, it appears that, overdrafts / loans are used by Real
Estate Funds, Hedge Funds and, to a more limited extent, by Funds of Hedge Funds,
over the 10% limit authorized for UCITS funds. Overdrafts / loans are a way to increase
the market exposure and hence to leverage the portfolio. Since leverage amplifies the
impact of market movements and may result in larger losses, it is seen as a source of
risk. The difference in terms of liquidity requirements for UCITS and non-UCITS,
together with the associated risks will be further described in the following section.
- Derivatives: they are present in hedge funds, Funds of Hedge Funds and in the “other”
category and they may also lead to increase the leverage (“instrument leverage” since it
arises through off-balance sheet derivatives positions). Derivatives appear however to
have a lower utilisation rate than in our UCITS sample. The use of derivatives will be
further analyzed below, to have a clearer idea in how non-harmonized funds invest in
derivatives.
European Commission
DG Internal Market
 61
The following schedule gives an indication of the frequency of use in the surveyed
population of the respective derivative types generally, irrespective of the fund categories.
Percentage of non-UCITS funds using the following derivatives products (2002-2006)
2002 2003 2004 2005 2006 Average
FET 27% 30% 31% 34% 27% 30%
Futures 4% 17% 18% 25% 25% 18%
Repurchases/Reverse Repurchases 17% 6% 11% 8% 6% 10%
Options 4% 5% 7% 8% 6% 6%
Swap 2% 5% 6% 7% 4% 5%
CFD 1% 1% 1% 1%
Swaption - - - - - -
When comparing these statistics it has to be noted that, whereas the most frequently used
derivative type by UCITS funds is the Futures contract, non-UCITS funds tend to primarily
use forward exchange rate contracts, whereas futures are in second position. Other
derivative types are used in less significant proportions and swaption are even not present
in the surveyed sample.
The following charts present the level of "sophistication", in reference to the number of
lines of derivatives observed, of non-harmonised funds; they correspond to the proportion
of funds present compared to the total population within each category.
2002
0%
20%
40%
60%
80%
100%
FoHF Hedge Fund Other Private Equity Real Estate
0 <50 over 200
2004
0%
20%
40%
60%
80%
100%
FoHF Hedge Fund Other Private Equity Real Estate
0 <50 50 to 200
2003
0%
20%
40%
60%
80%
100%
FoHF Hedge Fund Other Private Equity Real Estate
0 <50 50 to 200
2005
0%
20%
40%
60%
80%
100%
FoHF Hedge Fund Other Private Equity Real Estate
0 <50 50 to 200
European Commission
DG Internal Market
 62
2006
0%
20%
40%
60%
80%
100%
FoHF Hedge Fund Other Private Equity Real Estate
0 <50 50 to 200
These charts evidence a stable use of derivatives by non-UCITS funds and no evolution
suggesting an increasing in the level of sophistication is noted as the proportion of funds
disclosing more than 50 lines of derivatives remain limited to hedge funds and the
proportion within this category is also constant.
In addition to this, information that needs to be captured is the number of derivative types
used in average by non-UCITS funds, depending on the category they belong to.
Average number of derivative types held by non-UCITS funds
(2002-2006)
0
0.5
1
1.5
2
2.5
FoHF Hedge Fund Other Private Equity Real Estate
2002
2003
2004
2005
2006
These results indicate that the number of derivative type used is less important than in the
UCITS environment and that most non-harmonised fund types (except for hedge funds and
others) use only one derivative type, mainly FET, which are used for hedging purposes in
many instances; no derivatives have even been observed for the years 2003 and 2005 in
surveyed private equity funds. Hedge Funds are the investment vehicles that use a wider
range of derivatives in a consistent way over the respective years, followed by funds
belonging to the “other” category.
European Commission
DG Internal Market
 63
This trend is supported by the percentage of funds within each category, which use the
respective derivative types; as presented in the tables below for the years 2002 to 2006.
2002
FET Futures Options
Repurchases/Reverse
Repurchases
Swap CFD
FoHF 50% 0% 0% 5% 0% 0%
Hedge Fund 20% 20% 20% 10% 0% 0%
Other 0% 0% 0% 33% 33% 0%
Private Equity 0% 0% 0% 33% 0% 0%
Real Estate 0% 0% 0% 43% 0% 0%
2003
FET Futures Options
Repurchases/Reverse
Repurchases
Swap CFD
FoHF 52% 3% 0% 0% 3% 0%
Hedge Fund 22% 56% 17% 6% 6% 0%
Other 0% 0% 0% 33% 33% 0%
Private Equity 0% 0% 0% 0% 0% 0%
Real Estate 0% 0% 0% 25% 0% 0%
2004
FET Futures Options
Repurchases/Reverse
Repurchases
Swap CFD
FoHF 53% 0% 0% 3% 0% 0%
Hedge Fund 24% 60% 20% 12% 16% 4%
Other 20% 20% 0% 40% 20% 0%
Private Equity 0% 0% 0% 17% 0% 0%
Real Estate 7% 0% 7% 20% 0% 0%
2005
FET Futures Options
Repurchases/Reverse
Repurchases
Swap CFD
FoHF 62% 4% 0% 2% 4% 0%
Hedge Fund 25% 75% 25% 8% 17% 3%
Other 14% 14% 0% 29% 0% 0%
Private Equity 0% 0% 0% 0% 0% 0%
Real Estate 8% 0% 4% 17% 4% 0%
2006
FET Futures Options
Repurchases/Reverse
Repurchases
Swap CFD
FoHF 45% 6% 4% 0% 1% 0%
Hedge Fund 19% 65% 10% 13% 8% 4%
Other 14% 14% 0% 14% 0% 0%
Private Equity 0% 14% 0% 14% 0% 0%
Real Estate 4% 0% 4% 0% 4% 0%
The results of the analysis conducted within this section emphasise the limited and stable
use of derivatives made by non-harmonised funds, compared to UCITS funds.
European Commission
DG Internal Market
 64
In the case of Fund of Hedge Funds, Private Equity Funds and Real Estate Funds, which
invest in limited and specific asset types, observations tend to evidence that derivatives are
used for hedging purposes and not for performance enhancement.
 Funds of Hedge Funds for instance primarily invest in forward exchange contracts
(FETs), which may be used in order to protect the fund against any currency fluctuation
between its own accounting currency and the currency of its investments. Frequency of
use of other derivative types appears to be anecdotic.
 Private Equity globally make a very limited use of derivatives as only 16% of such
funds used derivatives in 2006, which is the year in which the highest results were
recorded in that respect. In addition, they appear to be used on an ancillary basis and not
as a key element of the investment strategy, in regard to the quantity and weight of the
instruments present in the portfolio.
 Real Estate Funds equally invest in FETs, Options and Swaps, but in a very limited
proportion. This may be explained by the specific nature of their investments, which
imply a medium- to long-term holding period, which would make hedging costly or
inefficient over such a long period of time. The purpose of the use of FETs would
appear to be to hedge other liquid assets (not the principal investment in real estate) and
swap to protect against changes in interest rates linked to borrowings made by the fund.
Here as well this does not correspond to products used by a significant proportion of the
surveyed funds.
Hedge Funds managers develop some alternative investment strategies that may involve
any kind of assets. In those strategies, the use of derivatives can increase exposure to some
assets and reduce transaction costs, compared to direct investment in securities for
instance. That being said, in this environment, portfolio managers develop some alternative
strategies, which aim at detecting specific market opportunities on new developing
markets, or implementing some innovative asset management models. In this context,
derivatives are effectively used, as evidenced in our survey. However, their use is less
widespread than in the UCITS environment, and this can be explained by the following:
 If UCITS funds may face daily subscriptions or redemptions, many hedge funds appear
to be soft closed by large redemption fees. In that context, whereas UCITS may need to
adjust their investment positions almost on a daily basis, hedge funds benefit from a
more stable environment. Whereas the use of derivatives enables UCITS to reduce
transaction costs resulting from the requirement to adapt investments to capital
movement, hedge funds may prefer direct investment in specific assets.
 Investment restrictions applicable to hedge funds are quite limited, or defined on a
contractual basis, in their prospectus. As a consequence, these can invest in almost any
asset type as long as it corresponds to their investment strategy. They do not need to
invest in derivatives in order to gain exposure on an asset, which would not be eligible
as direct investment, as is the case for UCITS funds for commodities or hedge fund for
instance.
The analysis of the portfolio composition in 2006 and changes thereto highlights some
significant changes in terms of asset composition, within the UCITS universe, whereas
some asset types remain present in the non-harmonised environment only.
European Commission
DG Internal Market
 65
An analysis of these investments and their associated risks will enable us to better assess
how risks arising from UCITS products have evolved over the past 5 years. Also, this will
give us a clearer picture of the differences in terms of risk profile between UCITS and non-
harmonised funds, not limited to the market risk already analysed in section 4.
5.2 Risks associated with changes to the portfolio composition
When considering the overall risk profile of funds, two fundamental types of risk must be
taken into account i.e. investment risk and operational risk. Investment risk relates to the
uncertainty about the future benefits to be realized from an investment. It principally
includes the market risk, usually measured as the historical volatility of returns and already
analysed in section 4 above, but also the liquidity risk and the counterparty risk. As regards
operational risk, it is defined as the risk of loss resulting from inadequate or failed internal
processes, people and systems or from external events
4
.
In the light of these definitions, we will assess the risks associated with new investment
types introduced in the UCITS universe and those associated with the non-harmonised
environment, without taking into account the market risk at this stage.
This analysis will be complemented with the answers given by asset managers during the
risk management survey we conducted as part of this study in order to consider their
perception of the risks funds are exposed to. More specifically, the survey asked
participants about the processes and tools that have been developed within their
organization and how appropriate they were to the measurement of specific risks. This
section will only cover the main findings from the survey, the detailed answers and
responses are presented in APPENDIX H (p.129).
5.2.1 Changes to the UCITS environment
The analysis of the portfolio composition of UCITS and changes thereto shows 2 major
trends: the proportion of target funds in the population has increased significantly, and the
use of derivatives also seems to be more diversified and in slightly larger proportion as
from 2005. Whereas in the former case, this derives directly from the new investment
powers, in the latter, it is mainly due to the global developments in the derivatives market,
where OTC derivatives transactions are booming and to a lesser extent to the new
possibilities introduced by UCITS III.
a) Investment in target funds
UCITS III provides for the creation of fund-of-fund structures, which was not possible in
the context of UCITS I. In the analysis of the portfolio composition, we note that
underlying funds are mainly domiciled in Europe and benefit from the European passport
(i.e. UCITS funds).
4
Definition of operational risk by the International Convergence of Capital Measurement and Capital
Standards known as Basel II
European Commission
DG Internal Market
 66
Based on the limited information included in the financial statements, we did not get any
relevant view on the weight of listed vs. unlisted target funds in the portfolio of the
surveyed funds. However, whether or not UCITS funds are listed does not appear to be a
key criterion to evaluate their liquidity, as any shareholder needing some cash will ask the
fund to redeem its shares without needing to go to the market.
UCITS need to be able to meet any redemption from a shareholder upon request. When
investing in a UCITS, a fund ensures a certain liquidity level of its investments. In that
perspective, the use of this new investment possibility does not appear to have had any
impact on the liquidity risk of UCITS investing in other UCITS, even if the current sub-
prime crisis leads us to mitigate this statement. Actually, some specific market events may
lead funds to suspend NAV publication and possibility to redeem over a certain period of
time. However, this liquidity risk is not deriving from the investment vehicle used itself
(ie fund) but from the underlying investment/market liquidity. When investing in a fund, a
fund is exposed to the liquidity risk associated with investment/market of the target fund
directly (as it would be when investing directly in same asset type) but does not increase its
own liquidity risk.
The operational risks arising from the introduction of this new investment power are
mainly linked to the daily valuation. Prices of such investment are not always published
through main pricing providers, but need to be retrieved from the administrative agent of
the target fund, who is responsible for the publication of the NAV per share information. In
terms of organisation, this implies that a specific process needs to be implemented to
collect NAV data for each new investment fund present in one portfolio. Any delay of
failure in the daily collection of the price of the target fund can have an impact on the
resulting NAV of the investing fund; in this circumstance, subscription / redemption orders
processed based on an incorrect NAV can be detrimental to the investors and to the Fund
as a whole, as the prices applied to these capital movements will not correspond to the
actual value of underlying investments. In addition, the quality of the price received can be
difficult to challenge by administrative agents in charge of asset valuation. In the context
where market practice is to obtain a price for any security from two different sources, or
develop internal tools to challenge the price received from third parties, this is more
difficult to implement in case of target fund pricing. In that circumstance, incorrect pricing
of investment made in target funds may be more difficult to assess. UCITS funds are
supervised and need to be audited; this provides an annual assurance on the processes in
place to value the target funds but does not give any daily certification on the issue price
received from the administrative agent.
b) Analysis of the impact of a wider use of OTC derivatives
In comparison with UCITS I, UCITS III moved from limits on use of derivatives based on
the commitment, to a limit set in terms of global exposure or risk. If this change enabled
some new investment opportunities, through a wider use of derivatives in UCITS funds,
this development is also in line with a global market trend where OTC derivatives have
become an increasing and significant proportion of the global derivatives market. As an
example, the notional amounts outstanding of the global credit default swap (CDS) market
were multiplied by 2.8 between the end of June 2005 and the end of December 2006,
according to the Bank for International Settlements (BIS).
European Commission
DG Internal Market
 67
The increasing use of OTC derivatives is also raising some specific risks associated with
OTC transactions, i.e. counterparty (due to the absence of a clearing house), but also
liquidity risk and leverage risk. The requirement in terms of quality assessment of the
issuer set in the directives aims to mitigate the counterparty risk.
Concerning the liquidity risk, OTC products are not designed to be sold on the market
before maturity. As a consequence to close a contract, the investment managers need to
have a counterparty which agrees to close the position by taking the adverse position.
Unfavourable market conditions can translate into high costs or losses for a fund.
The leverage risk seems to be limited as the portfolio composition analysis showed that in
practice, the increase observed in the use of derivatives has not generated more leverage.
The specificity of OTC products is that they are not standardized because they are tailored
to meet the specific needs of investment managers. On the one hand, their design is really
flexible and can adapt to the specific needs of portfolio managers as regards underlying
assets and contract size. On the other hand, these features seem to imply specific
operational risks associated with this kind of investment. As any product is potentially
unique, this can result from a misunderstanding by the administrative agent or compliance
department of the product acquired by portfolio managers.
The valuation of such products, which are not traded on a regulated market is also more
complex, as they need to be independently valued, based on potentially complex valuation
models that administrative agents cannot develop internally; the resulting valuation is
difficult to challenge, based on an alternative pricing source. The specific features of these
products can lead to errors in the computation of the net asset value of the funds. In the
context of UCITS where daily subscriptions and redemptions are possible and expected,
any error can result in a material misstatement of the NAV per share used to evaluate the
shares subscribed for and redeemed and impact a large number of shareholders.
As innovative and sophisticated products are developed, the risk of selling inappropriate
products to clients who know too little about associated risks may increase. The disclosure
in the offering documents and in the financial statements of the type of derivatives used
may be insufficient for retail investors to understand the risk associated with some
products or investment strategies and the level of exposure resulting from such
investments. In particular, some “guaranteed” products may be misleading as to the
guarantee associated with the products, as to whether the guarantee refers to the initial
investment, to a guaranteed performance, or only to lower sensitivity to any significant
unfavourable market developments.
However, the risk management survey and face-to-face interviews showed that asset
managers implement strong risk management and valuation procedures before investing in
such complex products. Another result from the portfolio composition analysis is the
limited number of funds using derivatives extensively.
European Commission
DG Internal Market
 68
In order to assess the impact of the operational risks associated with the current market
trends and changes in portfolio composition, we analyzed the data reported by Central
Administrations to the Luxembourg Supervisory Authority (CSSF), in the context of the
specific legal framework aiming at protecting investors against impacts of material NAV
calculation errors and non-compliance with investment restrictions. This is the only public
data available on that subject in the 9 countries of our sample.
c) Practical implications – Luxembourg example
The statistics reported by the CSSF in its annual activity reports give a bit more insight into
incidents reported to the Luxembourg Supervisory authority by Central Administrations.
The volume of NAV computation errors and investment breaches, together with their
financial implications, meaning amounts reimbursed to the funds and / or to their
shareholders are published in these annual reports. In these documents, the CSSF
distinguishes between issues relating to NAV calculations errors and those relating to
investments made which are not in compliance with the Fund's investment restrictions or
legal requirements (“investment breach”).
The analysis of these statistics shows an increase in the number of incidents of both natures
between 2002 and 2006, but which follows the trend of market growth. Concerning the
NAV calculation errors, the percentage of errors resulting from incorrect valuation of
futures or swaps remains unchanged over the period. The most significant development is
the increase in NAV errors resulting from accounting entries and from trading errors,
compensated for by a decrease in pricing errors. In terms of compensation granted to
impacted shareholders, it appears that the proportion of significant reimbursements to the
funds or shareholders has decreased over the period.
Even if there is no direct link possible between the new investment powers introduced,
their associated risks and the CSSF reporting framework, it has to be noted that in the
Luxembourg market the increase in the use of derivatives does not seem to have had any
negative impact on the administration and management of funds.
Additional details on CSSF statistics are enclosed in APPENDIX E (p.123).
If the investment opportunities of UCITS have expanded over the years, some investments,
however, remain dedicated to the non-harmonised environment. The following section will
consider these specific investment types and their associated risks in order to have a more
comprehensive analysis of their risk profiles.
European Commission
DG Internal Market
 69
5.2.2 Non-harmonised universe specificities
A key difference between UCITS and non-UCITS funds is the requirement for UCITS
funds to take into account their own liquidity requirements in order to be able to meet at
any time any redemption request from a shareholder, in accordance with article 37 of the
UCITS Directive. However, the sub-prime crisis showed that UCITS may face liquidity
issue and the good rating of some investments (Asset Backed Securities, Mortgage Backed
Securities, etc) is not an assessment of their liquidity. That being said, unlike UCITS funds,
non-UCITS funds can impose notice periods, lock-up periods, and gate provisions, which
reduce their liquidity constraints. If this can result in less liquidity from an investor’s point
of view, it enables non-harmonised funds to invest into specific asset classes and to keep
their level of cash to a minimum.
In practice, non-UCITS funds can invest into non-listed and illiquid assets such as real
estate and private equity. If those investments do not appear to have a distinctive behaviour
in terms of market risk, they have the following features however: they are more difficult
to value, it takes longer to buy or sell them, the time horizon and the risk concentration
they imply. Whereas the market value of securities exchanged on a stock exchange can be
easily assessed on a daily basis and is able to factor in the latest market or economic
developments, illiquid investments need to be valued by external experts, who sometimes
need a few weeks to assess the value of a specific asset. The timing difference implies that
the value of these illiquid investments may be few weeks late compared to more recent
market developments. In the context of funds where capital movement could take place at
any time, this could generate a significant risk of market timing, which can be higher than
for UCITS funds and need to be properly assessed and managed. In addition, when selling
a security, the risk that there is a difference between an expert valuation and the actual
market value of an asset is also higher than for funds investing in liquid assets. This
difference can result from a low liquidity of an investment that may not be easy to sell on
the market due to a lack of potential buyers. If this can result in a lower market value, it
can also imply some delays in realising those investments; these two drawbacks can
directly affect the shareholders of a fund, who have to bear the financial impact, but also
wait to actually retrieve their cash.
Another distinctive requirement for UCITS funds is to follow some strict rules concerning
risk diversification, in terms of counterparty risk (maximum exposure to one counterparty),
but also in terms of number of different securities held in the portfolio and their maximum
weight. This diversification requirement does not apply to non-UCITS vehicles, which
have less stringent rules. In case of default or sizeable decrease in the value of one
investment, this can have a significant financial impact for shareholders.
Another key observation from the portfolio composition analysis is that hedge funds
heavily invest in commodities, whereas exposure to a single commodity is still forbidden
for UCITS.
European Commission
DG Internal Market
 70
Even though we noted that UCITS funds use the fund-of-fund structures introduced by
UCITS III, we need to point out, however, that UCITS funds cannot invest in the same
fund types as non-harmonised funds. Under the UCITS directives, target funds must
comply with supervision and risk diversification principles similar to those applicable to
UCITS funds and with similar liquidity requirements for investors. The risk profile of
target funds therefore appears to be different for UCITS funds or funds of hedge funds.
The specific features limit the availability for UCITS funds of investments or investment
strategies listed above; even if, in terms of performance and volatility they appear to imply
a market risk comparable to UCITS investments (see section 4). If direct investment is not
possible, UCITS portfolio managers tend to gain exposure to this kind of investments
through the use of specific investment vehicles such as Real Estate Investment Trusts
(“REITs”) for real estate market for instance or derivatives for commodities or short sales.
REITs are often closed-ended structures, which are listed on a stock exchange. As a
consequence, if the valuation of such investments derives from the value of the underlying
investments, they also include market components and the impact of a bid and offer on the
stock exchange. Given these vehicles are in most instances traded on official stock
exchanges, the risks related to these vehicles valuation and liquidity of the investment
seem more appropriate for UCITS funds, than direct investment.
Concerning derivatives, UCITS funds can now get exposure to a basket of commodities by
investing in this kind of contract, with limited transaction costs.
Short-selling is commonly used by Hedge Funds but this technique may have a significant
impact on the portfolio’s performance since it exposes the fund to an unlimited risk. Unlike
Hedge Funds, short-sales risk for UCITS funds is limited by the restrictive use of
derivatives, and prohibition to enter into uncovered short sale transactions.
Concerning financial leverage, UCITS funds are not allowed to borrow money as opposed
to non-UCITS funds that can use lending facilities granted by prime brokers or banks as a
means to increase their market exposure and associated risks.
Capco looked at why some hedge funds failed (analysis enclosed in APPENDIX G
(p.127)) and made the connection with corresponding risks. A key result is that 38% of
failures were associated with investment risk only while 54% were due to operational
issues.
Risks are present in both UCITS and non-harmonized funds even if some specific items
have been identified for the second category. It is now important to consider how managers
assess these risks and related developments. More importantly, the tools they have
developed to monitor these risks and ensure appropriate investor protection will also be
discussed.
European Commission
DG Internal Market
 71
5.2.3 Managers’ point of view on risk management and on the changes of risk profiles
We conducted a risk management survey involving 55 asset managers. Additional details
concerning the respondents’ profiles and the features of their funds under management are
disclosed in APPENDIX H (p.129).
The purpose of the questionnaire was to identify how they perceived the changes to the
risk profile of the funds they have under management and to look at the tools and
procedures that have been developed in this context.
The results of the survey confirmed the moderate use of derivatives already pointed out in
our analysis of portfolio composition in section 5.1. 63% of respondents indicate a
moderate to very moderate use of derivatives in UCITS funds (53% for non-UCITS) and
10% no exposure for UCITS (7% for non-UCITS) at all within their respective range of
funds (APPENDIX H, graph 6 (p.133)). In addition, derivative exposure is mostly
motivated by hedging purposes (92%) and almost half the respondents also use derivatives
for trading or arbitrage purposes (APPENDIX H, graph 7 (p.133)). When considering the
type of derivatives used, if almost all respondents (i.e. over 80%) report using plain vanilla
options, forward contracts or swaps, a significant portion (close to or over 50%) also
presents more complex products such as structured products, credit linked notes or ABSs
in their portfolios (APPENDIX H, graph 8 (p.134)). In this context, it is interesting to
consider the risk management tools they have developed, together with their perception of
the efficiency / coverage of such tools.
As a preliminary remark, it is worth noting that for 48% of surveyed asset managers, their
Risk management unit has not changed significantly following the conversion to UCITS III
regulations. This is especially the case for those who have the largest risk management
team (APPENDIX H, graph 9 (p.134)). In addition to the Risk Management Committee,
69% also have a New Product Committee, which in all cases (100% of responses) provides
final approval on new products.
From a global point of view, the overall assessment made by surveyed managers is that all
potential risks to which a fund is subject are covered appropriately by the risk management
framework, especially Market Risk, Counterparty Risk and Operational Risk
(APPENDIX H, graph 17 (p.139)).
As highlighted in our analysis of portfolio composition, one of the main market
developments, especially concerning UCITS is an increased investment in OTC
derivatives. The following paragraphs will cover the risks and organisational implications
associated with this specific investment type.
Fund managers confirmed during face-to-face interviews that the potentially increased
leverage resulting from the use, or lack of use of, OTC derivatives is limited and is not
sought after by asset managers and investors.
European Commission
DG Internal Market
 72
Investing in OTC derivatives implies the creation of specific tools to properly monitor the
pricing of these products, but also other risks associated with these. These tools need to
match the specific characteristics of each derivative type and properly monitor the risks
associated with these; they can be complex and involve many parameters and complex
mathematical models. In the light of current market developments, a majority of
respondents (59%) confirmed their confidence in the ability of the system to deal with
simple derivatives, whereas more complex products seem to be far from being handled
completely. However, 57% of respondents feel certain that their system will be capable of
coping with new products and strategies when required (APPENDIX H, graph 19 (p.141)).
Concerning the specific risk associated with the pricing of derivatives, respondents do not
put too much confidence in one unique pricing source but prefer the use of counterparty
prices or internal pricing models (APPENDIX H, graph 24 (p.145)). Even if the pricing of
new investment products is reported as difficult for 54% of the UCITS fund managers and
45% of the non-UCITS fund managers surveyed, the valuation risk itself does not appear
to be significant for complex instruments (APPENDIX H, graph 26 and 27 p.146 and
147)). The main reason behind this statement is that valuation issues are often considered
carefully before accepting new investments. 59% of respondents believe that the ability to
independently price an instrument before trading it is a prerequisite. This percentage
climbs to 88% for asset managers of sophisticated UCITS. It appears that this is not always
a prerequisite for asset managers of non-harmonised funds. As evidenced in the context of
the subprime crisis, valuation risk is more significant with investments which are deemed
to be standard and where unexpected market changes cannot be properly monitored and
integrated into pricing models on time. In this specific situation, the improper monitoring
of the liquidity of this market led to significant pricing adjustments of some investments.
In this context, it is interesting to note that only 11% of asset managers consider that the
liquidity risk is fully or mostly covered by the VaR calculation, but 54% state that it
remains a sufficient risk measurement tool to cover all products and portfolio risks
(APPENDIX H, graph 32 (p.150)). A specific liquidity analysis for authorised products
does not appear to be conducted for 26% of respondents and is conducted partially for 41%
of them (APPENDIX H, graph 40 (p.153)).
In addition to the liquidity risk, the counterparty risk is logically measured by most asset
managers (79%), which may be justified by the fact that this is imposed by UCITS III
regulations. Counterparty risk measurement tends to be harmonised among asset managers
even if it is not harmonised by regulators as shown by the comparative Regulatory
Analysis. Asset Managers tend to adopt the most conservative approach when performing
measurements and use collateral agreements to reduce counterparty exposure for 81% of
respondents (APPENDIX H, graph 41 (p.154)).
While UCITS funds need to invest primarily in transferable securities, non-UCITS funds
can potentially invest into any kind of asset. These may not always be traded on regulated
markets, or through standardized processes, which increase operational risks funds are
exposed to when entering into such transactions. In order to operate in that environment,
managers and administrators therefore need to implement specific controls and procedures
in order to ensure that trade transactions are booked properly and that the investments are
valued appropriately.
European Commission
DG Internal Market
 73
This specificity and associated risks (such as liquidity risk, counterparty risk, and
operational risks) can lead to a significantly different risk profile for non-harmonized funds
compared to UCITS, which was not evident at first, based upon the analysis conducted in
section 4, which focused only market risk.
However, asset managers tend first to develop an appropriate risk management system
before implementing flexibility resulting from UCITS III.
In a changing environment, it is worth highlighting the latest trends and innovations
observed in the UCITS environment in terms of investment strategies that could not be
captured in the context of our analysis due to the constraint set in the selection of our
sample in terms of launch date, whereby funds needed to have a minimum track record of
one year to be selected.
5.3 Innovation in the UCITS Fund industry
The information provided in this section is not intended to provide a comprehensive list of
all innovative investment strategies introduced in the UCITS environment. These are only
examples of some developments, which highlight a kind of convergence with strategies
that were dedicated to the alternative investment fund universe a few years ago and appear
to be used increasingly today.
A first category can be defined by the search of absolute returns (to give positive returns in
almost all market conditions). These absolute return funds tend to define themselves in
terms of their risk level rather than their asset allocation. As an example, in
September 2006 a European promoter launched a UCITS fund which employs a Global
Tactical Asset Allocation whereby it takes advantage of relative valuation opportunities
due to market inefficiencies. For that purpose, the fund uses quantitative techniques to take
long and short exposure on investments through derivatives to reach an annualised target
return of 1 month Euribor + 4% over a three-to five-year period. A key element of the
investment strategy is the limit clearly set in the offering document in terms of VaR, in that
instance set as a one week VaR limit of 5%.
Another increasingly prevalent investment strategy is the “130/30” strategy, defined as the
one that goes long 130% of the Net Asset and goes short 30% of the same Net Asset,
keeping a beta of around 1 and maintaining full equity market exposure. As UCITS are not
authorized to short sell investments, this investment strategy is applied through the use of
derivatives.
Following the latest recommendations made by CESR in July 2007, Hedge Fund indices
could also be introduced as eligible assets. In that context, in August 2007, a French
promoter launched the first UCITS identified in the marketing documents as the “first
compliant hedge fund clone” under French law. The fund uses a hedge fund replication
strategy by using derivatives in equities, bonds and currencies to outperform the Hedge
Fund Research Index (HFRI).
European Commission
DG Internal Market
 74
Another innovative investment strategy is to track volatility as an asset class: if this was
originally limited to the sophisticated proprietary trading desks and hedge funds managers,
hidden assets (i.e. volatility, correlation, dispersion, etc) are becoming increasingly
accessible. In 2007, funds domiciled in the UK, which offer for the first time returns on
equity volatility through a UCITS III compliant structure, were launched. These funds
exploit the difference between implied and realised volatility by investing in equities
(about 20%), keeping a significant level of cash and swapping the equity performance with
the performance of arbitrage strategies.
In addition to innovative investment strategies, portfolio managers are also looking for new
investment themes, such as carbon emissions, targeting for instance companies with
hazardous emission-reducing techniques. According to the World Bank, the carbon trading
market tripled in 2006 to $30bn from $11bn in 2005 and trading volume is expected to
double this year and to double again in 2008. The global carbon trading market is expected
to reach $250bn by 2010. In this global market, New Carbon Finance, a carbon research
firm, estimated that global investment in carbon-focused funds jumped by 66% to $11.8bn
between November 2006 and April 2007. However, this is still a typical segment for hedge
funds, available in Luxembourg-based regulated non-UCITS.
 New generation of Exchange Traded Funds (“ETFs”)
Based on information published in the Feri data digest 2007, during the year 2006, 80 new
ETFs were launched, sales rose to €13.9bn and their market share of total European sales
almost doubled to 5%. With $106bn under management according to Morgan Stanley, the
European ETF market still lags behind the American ETF market which reaches $471bn.
Initially, the specific features of ETFs were the replication of indices through investment of
all securities composing an index, adjusting positions in accordance with benchmark
composition. Nowadays, the use of derivatives allows ETFs to track indices through equity
swaps rather than by acquiring securities. The first implication of this development is the
reduction in transaction costs for the funds concerned. For portfolio managers, this means
the introduction of new derivatives types and the possibility for UCITS funds to gain
exposure to a new and wider range of underlying assets such as commodities or private
equity for instance. Here are some examples of observed developments:
In the first semester of 2007, several promoters launched UCITS ETFs on Itraxx products
in Luxembourg in order to replicate the Benchmark Total return Index. In practice, the
funds are mainly invested into money market instruments and enter into total return swaps
or credit defaults to get exposure to credit risk, in line with the benchmark. Another
example is the creation of ETF products protecting investors against losses or finally the
creation, in January 2007, of ETFs developing investment strategies that take advantage of
a decline in a base index.
European Commission
DG Internal Market
 75
5.4 Conclusion on portfolio composition analysis
The analysis of the portfolio composition of UCITS and changes thereto shows an
effective use of wider investment powers introduced by the UCITS III directives. The
proportion of target funds in the population has increased significantly and the use of
derivatives seems also to be slightly more intensive, as funds including more than
200 derivative lines in their portfolio appear as from 2005 and keep developing in 2006 but
remain limited to a smaller proportion of the European Fund market. The increasing use of
derivatives, especially OTC products, is actually a key factor as regards risk since
UCITS III directives moved from a commitment approach to a VaR (value-at-risk)
approach.
Although derivatives do not necessarily increase market risk, they may affect other sources
of risks since they provide (to a limited extent) access to new markets, and make it possible
to take short-sales positions. While possible in principle, the use of derivatives to leverage
Fund exposure to market developments appears to be limited in the surveyed funds (only
1.6 on average and only for the most sophisticated funds of the sample, which were limited
in number).
Among derivatives, over-the-counter (OTC) instruments are those, which are the most
likely to affect counterparty risk and to pose operational issues, especially in terms of
valuation risk. The sophistication of UCITS funds may also lead to difficulties in
understanding and managing risks associated with products and strategies. This could
result in inadequate resources needed in terms of technology, processes or personnel that
are able to handle operating activities the fund is involved in.
However, it is important to note that in practice, the number of sophisticated funds is
limited. The use of derivatives is also limited even if an increase is observed, which has
also been confirmed by our risk management survey. In addition, concerning the use of
such products, fund managers tend to consider that leverage through derivatives is not
appropriate for funds dedicated to retail investors. The derivatives exposure is generally
motivated by hedging purposes (92%) and almost half of the respondents use derivatives
for trading or arbitrage purposes.
That being said, this survey and face-to-face interviews we conducted clearly show that
Asset Managers tend to develop strong risk management procedures before launching new
products or entering into complex products. Most of our respondents (68%) have a formal
risk management committee in place. All Asset Managers promoting sophisticated UCITS
have a Risk Management Committee. The Risk Management Committee is actively
involved in the approval process of new products strategies (81%) as well as in every
decision related to risk measurement methodology. A new products committee has been
established by the largest share of respondents (69%). The main reason for rejecting new
products refers to the impossibility of evaluating the new product (19%).
European Commission
DG Internal Market
 76
On an ongoing basis, the use of VAR is widespread, not only for sophisticated funds and is
indicated as an essential tool for Risk Management Units, including for Hedge funds and
Funds of Hedge Funds. The global assessment on this tool is that it properly measures most
risks funds are exposed to (sometimes in conjunction with some other ratios), with the
exception of liquidity risk, which does not appear to be specifically monitored by other
specific tools.
In comparison with the non-harmonised fund population, a limited number of investment
types are not present in the UCITS universe. However, a key differentiator when
comparing UCITS and non-UCITS is the liquidity risk with the requirement for UCITS
funds resulting from redemption requests, which can happen at any time. While UCITS
funds must comply with the eligible assets restrictions, non-UCITS funds can invest in
non-listed and illiquid assets such as real estate and private equity. Those illiquid assets are
also a major source of valuation risk and may raise additional risks in times of market
turmoil when it is not possible to sell them at their fair value. For 47% of respondent to our
survey, this results in a difference in risk arising from UCITS and non-UCITS funds.
Concerning the use of derivatives, the description of the UCITS population and how it has
developed would evidence a higher degree of sophistication in terms of level of use of
derivatives than non-harmonised funds, both in terms of derivatives types used but also in
terms of quantity.
A more detailed analysis of the use of derivatives and the impact of the sophistication of
UCITS will be provided in the following section, which looks at how regulators are
addressing the above-mentioned risks.
European Commission
DG Internal Market
 77
6. UCITS and Non-UCITS: Comparison of the regulations between
a sample of European Member States
The regulations applicable to UCITS and Non-UCITS funds are not implemented similarly
throughout the different Member States.
The analysis we conducted has highlighted some potential risks such as:
 Market risk in the performance/risk analysis;
 The risks linked to the increasing use of derivatives as shown by the portfolio
composition analysis (mainly valuation risk, risk linked to leverage, counterparty risk,
operational risk, etc);
 The risk that may be more specific to the Non-UCITS universe such as the liquidity
risk.
The risk management questionnaire and the face-to-face interviews give us a good
assessment of how those risks are identified, monitored and managed by asset managers.
This analysis focuses on how those risks are supervised by the regulators in the European
Member States sampled. The first part of our analysis is dedicated to UCITS.
Additional information on the comparative analysis is enclosed in APPENDIX I (p.161).
The second part of our analysis will be focusing on the limits which enable Non-UCITS to
have enough liquidity so as to face their redemption obligations. It will also focus on the
limits in terms of leverage and on the valuation of assets.
6.1 UCITS
The objective of the UCITS Product Directive was to allow for open-ended funds investing
in transferable securities to be subject to the same regulation in every Member State. It
attempted to expand the scope of permissible investments to keep pace with industry
changes, in particular the development and increasing use of derivatives. It was hoped that
this legislative uniformity established throughout Europe would help to achieve the EU’s
goal of a single market for financial services in Europe.
The reality differed somewhat from expectations due primarily to individual rules or
interpretations in each Member State. The purpose of this analysis is to highlight the
existing differences between the transposition and the interpretation of the Directive within
the different Member States:
European Commission
DG Internal Market
 78
We have looked at those topics directly linked to risks identified previously, to see whether
over the years, countries have evolved in their interpretation. The topics impacting risks
are the following ones:
 Risk-spreading ratios affecting the market risk;
 Distinction between sophisticated and non-sophisticated UCITS;
 Commitment approach for the use of derivatives;
 Risk-management approaches in term of VaR implementation;
 Leverage in UCITS;
 Counterparty risk for OTC derivatives;
 Short sales of derivative instruments and cover requirements;
 Valuation risk;
 Management companies, delegation of functions, use of passport.
6.1.1 Risk -spreading ratios affecting the market risk
Market risk is limited by the principle of risk spreading and by the concept of eligible
assets.
The main differences are observed for the so called “trash ratio” i.e. transferable securities
and money market instruments other than those referred to in paragraph (1) of the
article 19 (1) of the Directive. France is the most flexible country as even Hedge Funds and
Funds of Hedge Funds are allowed under certain conditions laid down by the French
regulator. Luxembourg has followed the French position but to a certain extent only.
France also allows the short selling of transferable securities to a maximum of 10% of the
total net asset value (when using repurchase agreement with transfer of ownership).
The only other European country allowing short selling of securities for a UCITS III is
Ireland where it has recently been permitted for so-called 130/30 funds.
6.1.2 Distinction between sophisticated and non-sophisticated UCITS
With the inclusion of financial derivatives as eligible assets for a UCITS fund, UCITS III
has paved the way for much more innovative investment strategies, generally much harder
to comprehend for the investor than under the previous regime. In 2004, the Commission’s
recommendations5 on the use of derivatives created the term “sophisticated UCITS”,
without however defining it, and imposed a risk management approach appropriate to that
sophisticated nature.
Only sophisticated UCITS would require the use of a Value-at-Risk (VaR) method,
whereas non-sophisticated UCITS could rely on a commitment method.
5
Commission recommendations of 27 April 2004 on the use of financial derivative instruments for
undertakings for collective investment in transferable securities (UCITS).
European Commission
DG Internal Market
 79
The following countries have defined the concept of sophisticated versus non-sophisticated
UCITS:
 Luxembourg
 France
 Germany
 Ireland
More precisely, the Luxembourg regulator has defined a sophisticated UCITS as a UCITS
using a high number of financial derivative instruments in the context of complex
investment strategies whereas a non sophisticated UCITS uses fewer derivative
instruments in the context of less complex strategies or for the purpose of hedging.
In France, the nature of derivatives used and/or the strategy followed by a specific UCITS
qualify the UCITS as sophisticated or not. However a non-sophisticated UCITS may use
complex derivatives without being considered sophisticated provided that the maximum
loss related to those complex derivatives does not exceed 10% of the UCITS assets.
In Germany, the qualified approach has to be applied as a rule. However, a UCITS may
apply a simplified approach as an alternative to the qualified approach if all market risks
within the investment firm can be adequately detected and assessed. A list of complex
derivatives requiring the use of a qualified approach has been defined.
In Ireland, sophisticated UCITS are defined as opposed to non-sophisticated UCITS, which
are UCITS using a limited number of simple derivative instruments for non-complex
hedging or investment strategies.
The following countries do not give a clear regulatory definition of the concept of
sophisticated versus non-sophisticated UCITS:
 Italy
 Spain
 The United Kingdom
 Slovakia
 Poland
In Italy, the Bank of Italy states that asset management companies that significantly invest
in derivative instruments must adopt suitable methods to measure and monitor risks arising
from the use of derivatives.
In Spain, the categories of derivatives and techniques used as well as the risks associated
with such categories of products will influence the regulatory conditions under which the
management company should adopt risk measurement systems.
In the United Kingdom, the market practice has defined some characteristics according to
which there is a presumption that a UCITS is sophisticated. Those characteristics are the
use of derivatives, which have a non-linear performance in relation to the underlying
assets, the use of OTC derivatives, the complexity of transactions, the use of derivatives as
part of the investment objective and the use of cover for derivative positions different from
the underlying of the derivative instruments.
European Commission
DG Internal Market
 80
No specific information is available for Poland and Slovakia.
6.1.3 Commitment approach for the use of derivatives
As a reminder, according to article 21 (3) of the Directive, the global exposure to
derivative instruments may not exceed 100% of the assets of the UCITS.
As mentioned above, the Commission recommendations of 27 April 2004 on the use of
financial derivative instruments for UCITS state that non-sophisticated UCITS could rely
on a commitment method to determine the global exposure to derivative instruments.
The following countries have foreseen the commitment approach in their domestic laws or
regulations:
 Luxembourg
 France
 Germany
 Ireland
 Slovakia
The Luxembourg and French regulators have provided additional information with regard
to the reinvestment of cash collateral received in the context of securities lending
transactions or repurchase agreements. For Luxembourg and France, these transactions will
have to be taken into consideration for the calculation of the overall risk.
The Spanish and the UK legislations do not specifically foresee the commitment approach
to calculate the market risk. However, as per market practice, the commitment approach is
used in this context.
No specific information is available for Italy and Poland.
The following countries have defined in their national laws or regulations the way the
commitment approach has to be applied for specific financial derivative instruments:
 Luxembourg
 Germany
 Ireland
The UK regulator did not include the calculation methods in the national laws or
regulations. However, calculation methods are given in a document issued by the UK
market associations for some categories of derivatives.
It is worth noting that the calculation methods suggested by the UK market associations are
different from the Irish and Luxembourg ones. Indeed, the UK guidelines foresee the use
of the market value, of the acquisition price or of the exercise price of the options
depending on the type of contract. Concerning contracts for difference, the market value of
the contract or the exercise price is suggested depending on the cash or physical settlement
of the contract and finally the unrealised result is suggested in the context of forward
contracts.
European Commission
DG Internal Market
 81
Non-sophisticated French UCITS using the linear method are calculating a “potential
amplification” being the commitment and which can be obtained by the multiplication of
the leverage effect by the value of the fund’s assets.
There is no list detailing the calculation method of the “potential amplification” by type of
products, however this information is provided by nature of underlying (equities and
interest rates).
In France, currency derivatives should not be taken into account in the calculation of the
global exposure of a UCITS to derivative instruments. This is very interesting in terms of
product structuring as asset managers have more investment powers resulting from the use
of derivatives.
Concerning Slovakia and Spain, there is no calculation method clearly defined in the
legislation.
No specific information is available for Italy and Poland.
6.1.4 Risk management approach in terms of VaR implementation
A VaR approach for non-sophisticated UCITS is provided in the domestic laws or
regulations of a few countries.
The following countries have specifically authorised the use of a VaR approach for non-
sophisticated UCITS in their domestic laws and regulations:
 Luxembourg
 France
 Germany
 Ireland
As far as the United Kingdom is concerned, the use of a VaR approach for non-
sophisticated funds is foreseen in the guidelines issued by UK market associations (IMA
and FOA).
Considering the simple portfolio structure and the limited use of derivative instruments in
Spanish UCITS, the CNMV has not defined the concept of VaR for UCITS in its
legislation.
No specific information is available for Italy, Slovakia and Poland.
The VaR approach for sophisticated UCITS has been clearly defined in the national laws
or regulations of:
 Luxembourg
 France
 Germany
 Ireland
 Italy
European Commission
DG Internal Market
 82
Luxembourg, France, Germany and Ireland have further defined the concepts of “Relative
VaR” and “Absolute VaR” as follows:
Luxembourg France Germany Ireland
Relative VaR
The VaR of the UCITS
divided by the VaR of the
reference benchmark or of
the reference portfolio
without derivatives
Maximum 2 Maximum 2 Maximum 2 Maximum 2
Absolute VaR
The VaR of the UCITS
capted as a percentage of
the net asset value
Maximum 20% 5% if no
benchmark up to
10% but basic rule
is relative VaR
Twice the VaR of
the benchmark
Maximum 5%
In Italy, the Bank of Italy makes reference to the guidance provided by the European
Commission on the matter. It states that asset management companies that significantly
invest in derivative instruments must adopt suitable methods to measure and monitor the
risks arising from the use of derivatives (such as VaR systems adjusted to the activity
conducted).
As far as the United Kingdom is concerned, the use of the VaR approach for sophisticated
funds is provided for in the guidelines issued by the UK market associations. However, no
specific limit has been determined.
It is important to underline that the absolute VaR threshold of 5% for Ireland is expressed
for a holding period of maximum one month. Consequently, we understand that this
absolute VaR threshold of 5% could potentially be set for a holding period of one day and
therefore it would correspond to an absolute VaR threshold of 22.36% for a holding period
of one month.
As regards stress tests, Luxembourg and Germany have laid down the requirement that
they must be carried out at least on a monthly basis.
No specific information is available for Spain, Slovakia and Poland.
d) VaR approach – parameters
The following table discloses the VaR parameters by country:
Countries Lux. France Germany Ireland Italy UK
Level of
confidence
99% 95% 99% 99% 99% 99%
Holding
period
1 month 7 days 10 business
days
Max 1
month
1 month 1 month
Historical data Min 1 year - Min 1 year Min 1 year Max 1 year Max 1 year
Those parameters are not specifically provided for under Polish and Slovak legislation
European Commission
DG Internal Market
 83
6.1.5 Leverage in the case of sophisticated UCITS
A leverage effect may be obtained through borrowings and/or through the use of derivative
instruments.
The Luxembourg and the French regulators have not provided for any specific limit in
relation to the leverage.
For Spain, Italy, Slovakia and Poland, the market practice limits the leverage which is not
considered appropriate for UCITS funds.
6.1.5.1 Borrowings
For Luxembourg and the UK, an investment company may borrow the equivalent of up to
10% of its net assets provided that the borrowing is on a temporary basis and not for
investment purposes. Please note the fact that a UCITS may not borrow for investment
purposes is clearly mentioned in the regulation of those two countries while it is not the
case in the other EU Member States.
6.1.5.2 Derivative instruments
For sophisticated UCITS, the leverage may be the indirect result of the Value at Risk
calculation. As the global exposure is no longer calculated with the commitment approach,
the potential leverage may exceed 2. In practice, some countries impose a maximum
leverage of 2 even if the UCITS calculates a Value at Risk.
For Germany and Ireland, leverage in the case of sophisticated UCITS is foreseen in the
national law as follows:
Germany Ireland
Leverage Maximum VaR
= two times the
VaR of the
benchmark
Maximum leverage
= 100% of the
NAV
As far as the United Kingdom is concerned, the UK market associations mention that
"funds might wish to use derivatives that would leverage a fund's exposure from
derivatives beyond 100% of NAV (e.g. credit default swaps) and this would not be
acceptable using a commitment basis. However, subject to the controls being in place, VaR
and associated techniques can be used to calculate the exposure, provided that the VaR
outcome does not exceed a percentage level as agreed with the fund's depositary."
European Commission
DG Internal Market
 84
6.1.6 Counterparty risk
6.1.6.1 Calculation of counterparty risk as per article 22 (1) & (2) of the Directive
“The risk exposure to a counterparty of a UCITS in an OTC derivative transaction may not
exceed:
- 10 % of its assets when the counterpart is a credit institution referred to in Article
19(1)(f), or
- 5 % of its assets, in other cases.
Notwithstanding the individual limits laid down in paragraph 1, a UCITS may not
combine:
- Investments in transferable securities or money market instruments issued by,
- Deposits made with, and/or
- Exposures arising from OTC derivative transactions undertaken with a single body in
excess of 20 % of its assets.”
The Commission’s recommendations of 27 April 2004 on the use of financial derivative
instruments by undertakings for collective investment in transferable securities invite
Member States to use the standards laid down in Directive 2000/12/EC as a first reference.
The abovementioned document recommends that Member States require the assessment of
counterparty risk with regard to OTC derivatives in accordance with the marking-to-
market method laid down in Directive 2000/12/EC of the European Parliament and of the
Council (1), notwithstanding the need of appropriate pricing models when the market price
is not available. They should also require the use of the full credit equivalent approach laid
down in this Directive, including an add-on methodology to reflect the potential future
exposure.
The table below summarizes the methods used by Luxembourg, France, Germany, Ireland,
the United Kingdom and Italy to calculate counterparty risk:
Countries Lux. France Germany Ireland UK Italy
Positive replacement value - YES - - - -
Positive replacement value +
(notional * add-on)
- - - YES - -
Positive replacement value +
(positive replacement value* add-
on)
- - YES YES - -
Multiplier * (positive replacement
value + (notional * add-on))
YES - - - YES* YES
* Market practice
European Commission
DG Internal Market
 85
In Luxembourg, the counterparty multipliers are further defined as follows:
- 20% for EU credit institutions and investment companies or recognized credit
institutions and investment companies of non-EU countries; and
- 50% in the other cases.
The multiplier used in the United Kingdom is 20% for banks of the A zone (in the meaning
of the Capital Adequacy Ratio Directive).
On the other hand, the United Kingdom foresees that this methodology could not be
appropriate in all situations and therefore that other methods of assessing the counterparty
risk could be considered.
In Ireland, a percentage of the positive marked-to-market value of the financial derivative
instrument could be taken into account instead of the notional or underlying value.
Considering the simple portfolio structure and the limited use of derivative instruments in
Spanish UCITS, as it is the case for the VaR approach, the CNMV has not defined any
methodology to calculate the counterparty risk.
No specific information is available for Slovakia and Poland.
6.1.6.2 Netting
The Commission’s recommendations of 27 April 2004 on the use of financial derivative
instruments also provides for recognition of netting of counterparty risks.
The above mentioned document recommends that Member States allow UCITS to net their
OTC derivative positions vis-à-vis the same counterparty, provided that the netting
procedures comply with the conditions laid down in Directive 2000/12/EC and that they
are based on legally binding agreements.
Currently only Luxembourg, Germany and Ireland have specifically foreseen the netting of
counterparty risk on financial derivative instruments with the same counterparty provided
that a contractual netting agreement has been signed.
6.1.7 Short sales of derivative instruments and cover requirements
6.1.7.1 Prohibition of short sales according to article 42 of the European Directive
“Neither:
- an investment company, nor
- a management company or depositary acting on behalf of a unit trust
may carry out uncovered sales of transferable securities, money market instruments or
other financial instruments referred to in Article 19(1)(e), (g) and (h)”.
European Commission
DG Internal Market
 86
Luxembourg, Ireland, the United Kingdom and Slovakia authorise the short sales of
financial derivative instruments provided that they are adequately covered.
In Poland and Germany, short sales are forbidden. However, in Germany, derivatives
which provide an option for cash settlement and which do not create a physical delivery
obligation are not affected by this short selling prohibition.
The prohibition of short sales of financial derivative instruments is not specifically
provided for under French legislation.
The Spanish regulator does not authorise any UCITS to have short positions on financial
derivative instruments.
No specific information is available for Italy.
6.1.7.2 Cover requirements
The Commission’s recommendations of 27 April 2004 provide a definition of the cover
rules depending on the settlement characteristics of the financial derivative instruments, i.e.
physical or cash settlement.
According to these recommendations, UCITS are required to cover short positions on
financial derivative instruments as follows:
 If there is no cash-settlement: by the holding of the instrument underlying the derivative
or by other liquid assets if the underlying financial instrument is highly liquid and if the
additional market risk which is associated with that type of transaction is adequately
measured;
 If there is a cash-settlement: by holding cash, liquid debt instruments with appropriate
safeguards or by other highly liquid assets correlated to the underlying and subject to
appropriate safeguards.
The following countries have transposed this recommendation in their national regulations:
 Luxembourg
 Germany
 The United Kingdom
 Ireland
 Slovakia
However, two countries have introduced exceptions or precisions as described below:
 The German regulator has not adopted the possibility to have other liquid assets as
cover in the case of a transaction with physical settlement;
 The United Kingdom has not clearly defined the assets offering an appropriate cover but
has specified the appropriate level of cover. The use of cash receivable within one
month as cover for short derivative positions is also authorised.
European Commission
DG Internal Market
 87
In Ireland, a similar approach has been taken with small adjustments.
 The cash-settled financial derivative instruments must be covered at all times by liquid
assets, which are defined as money market instruments and transferable securities that
can be repurchased, redeemed or sold at limited costs, in terms of low fees and narrow
bid/offer spread, and with very short settlement delay.
 The financial derivative instruments with physical delivery must be covered at all times
by the holding of the underlying asset. The possibility to cover short positions by other
liquid assets is only foreseen for highly liquid fixed income securities and the specific
derivative should be addressed in the risk management process and details should be
provided in the prospectus.
Not applicable for Spain.
No specific information is available for France, Italy and Poland.
6.1.8 Valuation of OTC derivative instruments
In its articles 19 (1) (g) and 21 (1), the European Product Directive 2001/108/EC states that
OTC financial derivative instruments should be subject to reliable and verifiable valuation
on a daily basis and that this valuation should be independent.
According to the Luxembourg regulator, reliable and verifiable valuation is a valuation
made by the UCITS at the fair value and which is not only based on counterparty's market
values but which:
 Is based on a current market value or if not available on a valuation model using a
recognised and appropriate methodology; and
 Is checked at an appropriate frequency by an independent third party or by a unit inside
the UCITS but independent of the asset management.
In Ireland, the following guidelines are applicable:
 Financial derivative instruments shall be valued by a method clearly defined in the trust
deed, the deed of constitution or the articles of association;
 The UCITS is satisfied that the counterparty will value the transaction at least on a daily
basis and will close out the transaction at any time at the request of the UCITS at fair
value;
 The UCITS has implemented systems to ensure that valuations of OTC derivatives are
reliable. An independent check of the valuation provided by the counterparty should be
made at least weekly basis;
 Should an alternative valuation be used, independent from the counterparty price, this
valuation must be reconciled on a monthly basis to the counterparty valuation.
The French regulator gives information only for credit derivatives, which should be valued
on a daily basis with an independent check at least on a monthly basis.
European Commission
DG Internal Market
 88
The German law refers to the valuation of OTC derivatives as follows: "If no market price
is available for an OTC transaction, the price has to be determined in a comprehensible
manner on each valuation date."
 OTC transactions for which no market price is available may be executed only if the
management company is provided with appropriate assessment methods;
 Such assessment methods have to be documented in a comprehensible manner.
In the United Kingdom, the following guidelines could be applicable:
 The terms of the transaction in derivatives are approved only if, before a transaction is
entered into, the depositary is satisfied that the counterparty has agreed with the ICVC
or the authorised fund manager to provide a reliable and verifiable valuation in respect
of that transaction at least daily and at any time at the request of the ICVC or authorised
fund manager.
 A transaction in an OTC derivative under COLL 5.2.20 R (1) (b) must be capable of
valuation; a transaction in derivatives is possible only if the authorised fund manager
has taken reasonable care to determine the valuation of the derivative throughout its life
(if the transaction is entered into) and if it is able to value the investment concerned with
reasonable accuracy;
- on the basis of the pricing model which has been agreed between the authorised fund
manager and the depositary; or
- on some other reliable basis reflecting an up-to-date market value which has been so
agreed.
Again, considering the simple portfolio structure and the limited use of derivative
instruments in Spanish UCITS, the CNMV has not defined any methodology to value over-
the-counter derivative instruments in its circulars.
No specific information is available for Italy and Poland and Slovakia.
6.1.9 Management companies and delegation of functions
UCITS III has introduced the concept of management companies “with substance”,
meaning entities with specific capital requirements, and a need for qualified and reputable
managers, internal controls and sound administrative procedures. These management
companies see their activity as regards UCITS clearly defined (they are meant to do
“Collective Portfolio Management” or “CPM”) and even extended, for those choosing to
perform “Discretionary Portfolio Management” and ancillary, non-core activities. A
“passport” to perform all of these existing or new services within the EU borders has been
granted.
While already in place in certain countries (German or Austrian
“Kapitalanlagegesellschaften”, with a mandatory banking status and a share capital of 2.5
Mio EUR) this was a more challenging adaptation for other European jurisdictions, among
which Luxembourg and Ireland.
We have tried to get an update on the management companies’ activities, the scope of their
passport and of the way countries implemented parts of the new “substance requirements”.
European Commission
DG Internal Market
 89
We also wanted to know whether, akin to Luxembourg or Ireland, other countries would
accept extended delegation arrangements.
All Member States have seized the opportunity offered by the Profession Directive and
allowed delegation of one or more of their functions to third parties, provided that certain
conditions are met: the delegation cannot affect the management company’s liability for
the tasks delegated, it may not generate conflicts of interests (e.g. if the asset management
were delegated to the custodian) and it may not deprive the management company of the
substance, thereby making it a letter-box entity.
Many countries do allow for an extensive delegation of central administration functions
(fund accounting, NAV computation, register holding and TA function…), certain even
abroad, but only the UK goes as far as allowing a full cross-border outsourcing of these
tasks.
Asset management is often delegated by the management company to specialized asset
managers, within or outside of the promoter’s group. In some countries (Sweden, Spain,
Denmark) such delegation of asset management remains however essentially “national”,
while in others (Ireland, Luxembourg, Germany, Belgium) a cross-border delegation is the
norm. Despite UCITS III rules on delegation, some countries (France, Italy) still cling to
the principle of having “some” asset management performed by the management company
itself: in France for example, it would not be possible for a management company
authorised to manage equity funds to fully delegate the asset management of such asset
class. In Italy, the management company must keep certain strategic investment decisions
– stock picking can be delegated, not the asset allocation. In both countries, risk
management cannot be delegated abroad – even the actual “doing” must be made locally,
which is quite different from the prevailing situation in Luxembourg and Ireland.
The last element of CPM, i.e. “marketing”, is by nature generally delegated in the context
of open-architecture distribution or to group distributors, and often outside of the fund’s
country of origin.
Discretionary portfolio management, the service inherited from the ISD and now available
to management companies is performed in many UCITS III German, Austrian, French and
even quite a few Luxembourg UCITS III management companies. In all but one of the
surveyed countries, only the authorisation to provide DPM services also triggers the right
to perform the “non-core” services mentioned by the Profession Directive, i.e. investment
advice and safekeeping/administration of funds’ shares. French “type 1” UCITS III
Management Companies (those with a UCITS licence) can and even do perform non-core
activities such as investment advice, without being authorised to perform DPM services!
Nevertheless, if management companies cannot operate funds based in other member
states, a growing number of management companies (principally major players) create
branches abroad to provide distribution services, asset management or discretionary
portfolio management. This is the case for German KAG, which use their passport to
service Luxembourg funds. French “sociétés de gestion” also use their passport when they
provide asset management services to foreign funds.
European Commission
DG Internal Market
 90
Luxembourg management companies have passported themselves into Austria, France,
Germany, Italy, Netherlands, Spain, Sweden, UK, either by creating branches or by using
the “free provision of services” route. The passport is however rarely used for
administration services.
Finally for Luxembourg UCITS sold to the public, the CSSF requires the identification of a
so-called “promoter”. This concept does not exist in the regulatory framework; this is an
administrative practice from the CSSF.
The promoter is the entity which is at the origin of the UCITS and which determines the
orientations of the UCITS; this is the reason why the CSSF typically asks that the promoter
has majority representation on the board of directors of the SICAV/SICAF // of the
management company of the FCP (UCIs can be set up either as a SICAV/F or FCP in
Luxembourg).
The promoter has to fulfil certain requirements in terms of nature (no private individual
accepted), competence (basically, it must be a professional in the financial sector) and with
financial surface, so that it can deal with investor compensation if needed.
6.2 Non-UCITS
Non-UCITS are not submitted to any European regulations, therefore each Member State
has the possibility to implement its own regulatory framework.
The analysis conducted on a sample of Member States has shown interesting differences in
terms of liquidity requirements, limits on authorised leverage level and assets valuation
depending on the type of Non-UCITS, i.e. hedge funds or real estate funds.
6.2.1 Liquidity risk
Investment limits are defined in order to ensure that investments are sufficiently liquid and
diversified.
6.2.1.1 Hedge funds
In Luxembourg, diversification limits are applicable to transferable securities, investment
in other UCIs, margin deposits, premiums paid and commitment. In the context of the use
of financial derivative instruments, a liquidity reserve should be kept.
As it is the case in Luxembourg, the Irish and French hedge funds, the UK hedge funds set
up under the regime of Non-UCITS Retail Schemes and the German funds of hedge funds
may potentially encounter liquidity issues even if diversification limits are foreseen.
There are no specific guidelines for UK hedge funds set up under the regime of Qualified
Investor Schemes (QIS) and for German hedge funds.
In Spain and Italy, no diversification limits are foreseen, therefore, for example, a fund of
hedge funds may technically invest 100% of its NAV in one single hedge fund.
European Commission
DG Internal Market
 91
6.2.1.2 Real estate funds
The Luxembourg, French, German and Irish real estate funds as well as the UK real estate
funds set up under the regime of Non-UCITS Retail Schemes are submitted to
diversification limits when investing in real estate assets, however given the nature of the
assets, the liquidity of the UCI’s portfolio is not ensured.
We would like to draw your attention to the fact that in Germany a new law allows the
setting up of a Real Estate Investment Trust in the form of stock corporations (AG) with
publicly traded shares.
In Ireland, there is no restriction on the amount of cash or short term securities to be held
for the purpose of meeting the redemption requirements.
The Spanish regulator requests a diversification of the assets without imposing any limits
in the legislation.
In Italy, the Italian real estate funds are set up as closed-end funds and are submitted to
diversification limits.
6.2.2 Leverage
Leverage may be obtained by the use of derivative instruments, short selling and by
borrowings for investment purposes. The below analysis demonstrates that the authorised
level of leverage may be quite different from one country to the other. Some countries limit
borrowings, short selling and the use of derivatives separately (Luxembourg). Other
countries tend to limit globally the leverage (France, Spain).
6.2.2.1 Hedge funds
In Luxembourg, although UCIs pursuing an alternative strategy have no obligation to
borrow, their investment policy may foresee the possibility to borrow on a permanent basis
for investment purposes. This borrowing authorisation is limited to 200% of the net assets,
which is equivalent to a leverage of 300% through borrowing but this limit may be raised
up to 400% of the net assets (leverage of 500%) provided that the regulator approval is
obtained, and that the investment strategy be of the market neutral type.
Germany has defined the concept of hedge fund through either the leverage effect obtained
by the use of unrestricted borrowings and/or of derivative instruments, either through the
use of short sales. No limit has been defined.
Concerning German funds of hedge funds, leverage is forbidden.
In Ireland, a scheme is permitted to engage, to a limited extent in leverage through the use
of techniques and instruments for the purpose of efficient portfolio management.
The net maximum potential exposure created by the use of techniques and instruments or
created through borrowing or through both of these together, shall not exceed 25% of the
net asset value of the scheme.
European Commission
DG Internal Market
 92
However, this rule is not applicable in the case of hedge funds set up under the
qualification of Qualifying Investor Fund (QIF) or Professional Investor Fund (PIF).
In the United Kingdom, hedge funds set up:
- Under the regime of Non-UCITS Retail Schemes are authorised to borrow up to 10% of
their net asset value on a non temporary basis;
- Under the regime of QIS/Funds have the possibility to borrow up to 100% of their net
assets.
In France, borrowings are limited to 10% of the net assets with a further limit applicable to
the global commitment in relation to derivative forward instruments, borrowing and
temporary sales or acquisitions of securities that may not exceed:
- 100% of the net assets for hedge funds set up as ARIA;
- 300% of the net assets for hedge funds set up as ARIALEL;
- 100% of the net assets for funds of alternative funds.
In Spain, a hedge fund is limited to leverage its net assets up to 5 times.
There is no specific information available for Italy.
6.2.2.2 Real estate funds
In Luxembourg, the aggregate of all borrowings made by a UCI may not exceed in average
50% of the valuation of all its properties.
In Ireland, the borrowing amount is limited to 25% of the value of the net assets.
The comments given in the context of Irish hedge funds in relation to the net maximum
potential exposure and in relation to the QIF and PIF are also applicable.
In the United Kingdom, the borrowing limit for real estate funds set up under the regime of
Non-UCITS Retail Schemes has been taken from the regulatory framework of UCITS,
which provides for a limited possibility of leverage.
For German REITs set up under the new REITs-law in form of a stock corporation (AG)
the minimum capital must be at least 45% of the real property.
The French regulator has imposed a double limit in terms of borrowings, i.e. the
borrowings may not exceed 50% of its real estate investments and the cash borrowings
may not exceed 10% of its other assets.
The leverage in Spanish real estate funds is limited or even not authorised as the
borrowings are limited to 10% of the net assets.
In Italy, the borrowing amount is limited to 60% of the value of the net assets.
European Commission
DG Internal Market
 93
6.2.3 Assets valuation
6.2.3.1 Hedge funds
In Luxembourg, France, Germany, Italy and Ireland, there are no specific regulations with
regards to the assets valuation for non-harmonised funds.
In Ireland, UCIs are requested to disclose in their prospectus the potential valuation
difficulties.
As opposed to Luxembourg and Ireland, the United Kingdom has specific regulations in
terms of valuation of OTC derivatives depending on the qualification of the fund as Non-
UCITS Retail Scheme or as QIS.
- Non-UCITS Retail Schemes: Criteria applicable to OTC derivatives in the context of
UCITS schemes apply.
- QIS: Valuation with reasonable accuracy on the basis of a pricing model or on some
other reliable basis reflecting an up-to-date market value.
The Spanish regulations foresee general guidelines in relation to the assets valuation, i.e.
daily valuation depending on the size and the nature of the asset, documentation and
control of the parameters used in the context of the valuation of derivative instruments.
6.2.3.2 Real estate funds
In Luxembourg, Germany, Ireland, Italy and the United Kingdom, independent property
valuers with a specific experience in the field of property valuation should be appointed for
the valuation of new acquisitions and for the regular valuation of the properties in
portfolio.
Whereas the valuation of all properties should be made at least on an annual basis in
Luxembourg, Germany and in the United Kingdom, the Irish regulator requires a valuation
at least twice a year. However, besides the annual full valuation with physical inspection, a
monthly valuation of the real estate assets on the basis of a review of the last full valuation
is required for the UK real estate funds.
The AMF in France requests two property valuers independent of each other and
responsible to draft a common report. However, no information has been given on the
frequency of the assets valuation.
6.3 Conclusion
This regulatory analysis showed that risks identified and listed at the beginning of this
section are subject to a high level of supervision but with some significant differences from
one country to another. Definition of sophisticated funds, Value at Risk implementation,
counterparty risk, calculation of commitment, potential leverage are not fully harmonised
across Europe. This is also the case for the functioning of management companies of
UCITS and delegation of functions.
Face to face interviews however, showed that such differences are not significant enough
to justify some regulatory arbitrages. For Non-UCITS, the liquidity risk does not seem to
be fully supervised and leverage limitations remain country specific.
European Commission
DG Internal Market
 94
7. General conclusion of the project
The section on performance/risk analysis does not show a significant change in market risk
following the implementation of UCITS III, on the basis of volatility. Market risk of non-
harmonized funds seems to be lower than for UCITS but this does not take into account the
liquidity issue which is often a downside for non-harmonised funds.
The section on portfolio composition did confirm an increased use of derivatives, including
OTC derivatives. This may result in additional operational risks linked to leverage,
valuation and counterparty risks. But, as explained above, such potential risks did not
impact the performance/risk profile for investors.
The risk management survey provides valuable insight into how such risks are addressed
by asset managers. It appears that risk management plays a key role in the process of
managing but also creating new funds. Asset Managers are usually prudent before
investing in new and complex instruments.
In particular, UCITS asset managers prefer to delay the launch of new products instead of
having a fund without proper risk management and valuation procedures, except for the
liquidity risk associated with certain investments which is not always included in the Value
at Risk calculation and in the assessment of new products. However, the ability to
independently price an instrument before trading is not always the rule for non-harmonised
funds.
The regulatory comparative analysis showed that more harmonization is still possible
concerning the concept of sophisticated funds, Value at Risk parameters or counterparty
risk. The level of precision of regulations in the sampled countries is different from one
country to another.
Some countries add more regulation to the European Directives when they have a
significant market of sophisticated funds with very active asset managers concerning the
use of derivatives.
In order to complete the project and in addition to the Risk Management survey, we
conducted two surveys with face-to-face interviews.
A first survey (detailed in APPENDIX J (p.162)) was undertaken, with a focus on
understanding the selection of fund domiciles by Asset Managers. Interviewees included
asset managers handling hedge funds, funds of hedge funds, private equity and real estate
funds.
We have conducted a second survey (detailed in APPENDIX G (p.127)) and discussed in
detail the process of transitioning to UCITS III, the consequences of UCITS III in terms of
product development and risk management, the convergence between sophisticated UCITS
and alternative investments, the leverage in the funds sold to European investors, and the
expectations concerning a future directive.
European Commission
DG Internal Market
 95
- Selection of fund domiciles
Criteria used by asset managers depend on the nature of funds proposed to investors.
UCITS asset managers never mentioned the level of flexibility in a given country for
selecting their fund domicile. This is explained by the current harmonization of rules
transposing UCITS III into the legislation of various European countries. Even if more
harmonization is still possible, as shown in the comparative regulatory analysis,
differences across Europe are not perceived as significant enough to justify the
domiciliation of funds. The criteria mentioned are the following:
- The acquaintance with the vehicle and demand by sales forces
- The time to market
- The flexibility of the tax system
- The reputation of the domicile
Respondents also place much importance on the existence of specific technical capabilities
such as cross-border transfer agents in Luxembourg or the experience in hedge fund
servicing in Ireland.
For hedge fund asset managers, the criteria are nearly the same, but the focus on efficient
service providers is more important as well as the scope of authorized investments.
As far as real estate asset managers are concerned, the flexibility of the tax system is at the
top of their list, followed by demand from investors and, to a lesser extent, the reputation
of the domicile and the time to market.
- Transposition to UCITS III
The transposition to UCITS III was a lengthy process. Asset Managers implemented the
new Directive as soon as possible to benefit from more eligible assets (mainly the creation
of funds of funds, cash and money market funds).
However, 90% of interviewees decided to wait for further regulatory clarifications before
creating sophisticated UCITS. It must be noted that most regulatory updates were issued in
2007 in the most active countries in terms of sophisticated funds like France and
Luxembourg. On the product side, we observe in 2007 an increase in the creation of highly
sophisticated funds. This sophistication is linked to the above-mentioned regulatory
clarifications but also to investments in more efficient and reliable risk management
procedures and systems.
- Impact of UCITS III
UCITS III had a very significant impact for all interviewees in terms of product
development. This is less the case, however, for the use of derivatives in France, Germany
and Luxembourg.
Derivatives were used extensively in those countries before UCITS III, but not always in
the context of harmonized funds. But, overall, interviewees believe that the use of OTC
derivatives has grown at a rate of 10% per annum since the implementation of UCITS III
and that such OTC derivatives allow for more flexibility in terms of product structuring.
European Commission
DG Internal Market
 96
The impact on Risk Management is more important, as developed in the risk management
survey. Strong Risk Management Units were already in place, but they had to adapt to
more complex products and cope with strong demand for financial innovation.
The final impact is the “substance issue” for management companies located in
Luxembourg and Ireland. Asset managers using UCITS III funds for cross-border
distribution purposes decided to set up their management companies very quickly, often
within 6 months after the transposition of the UCITS III Directives into domestic laws.
- Convergence between sophisticated UCITS and Alternative Investment funds.
All interviewees recognized that such a convergence is visible, as UCITS III allows the
structuring of funds replicating several alternative investment strategies with absolute
returns. We also observe the interest of Prime Broker, a service provider essential to Hedge
Fund managers, for sophisticated UCITS.
It is also important to note that UCITS vehicles are used extensively for distribution to
institutional investors. Some pension funds and insurance companies do require the
supervision and investor protection provided by UCITS III. However, when a UCITS
product is sold to retail investors only, risk features are different, in that they are more
suitable to investors having no risk appetite.
Finally, even if the regulation allows some UCITS products to implement alternative
investment strategies, asset managers always consider the suitability of their funds for
retail investors as well as the risk management process they need before launching such
funds.
- Leverage
In our sample, 80% of the asset managers consider that leverage is not appropriate for
retail investors, and it can be sold to institutional investors having a good understanding of
it.
In practice, leverage is very limited even if leverage opportunities resulting from
UCITS III are understood very well in France, Luxembourg and, to a lesser extent, in
Germany and the UK.
We should make a distinction, however, between leverage on the market (or β) and
leverage of some arbitrage opportunities (or the α). The leverage of the β is never
considered appropriate, except for some institutional investors. The leverage of the α may
be appropriate but with strong risk models and with appropriate disclosure to investors. But
even in that case, such leverage would be limited, as shown in the portfolio composition
analysis.
A last point mentioned by 60% of interviewees is that there is no demand for leverage by
investors.
European Commission
DG Internal Market
 97
- Expectations concerning a future Directive
All interviewees managing only UCITS are very satisfied with the flexibility provided
under UCITS III, but they would welcome additional powers in the following areas:
- Authorized investments in term loans
- Authorized investments in hedge funds
- Granting the European passport to feeder funds
However, all interviewees have expressed an interest for more harmonization concerning
the notification process and tax issues.
A strong interest has been expressed for a specific Directive with more eligible assets for
funds sold to institutional investors. A European passport for Hedge Funds in that context
is of interest, but it would have to be tailored to this type of fund.
Fund of hedge fund managers are also keen on a European passport on the basis of its
strong track record and with appropriate liquidity so that such funds can meet redemption
requests.
An important point concerning asset managers’ expectations is that no one among
interviewees is clearly interested in a future Directive based on risk management principles
rather than on investment restrictions and eligible assets.
They explained this point as follows:
- Even if risk principles are used increasingly, it is important to keep investment
restrictions used as guidelines by asset managers.
- Investment restrictions and eligible assets are objective criteria, easy to understand
for investors and often required by institutional investors.
- The concept of UCITS is linked to investor protection and it becomes a very strong
concept outside Europe. Investors in Asia and South America invest a lot in UCITS
products because of the quality of supervision, the excellent track record and
investment restrictions which are easy to understand by regulators in those countries.
Removing investment restrictions would be a threat to the UCITS concept in that
context.
- Finally, 70% of interviewees believe that the market is not ready; this includes some
asset managers but also most investors and regulators.
Nevertheless, asset managers are open to some risk management in the future, provided the
indicators are clear, recognized at a pan-European level and applied consistently across
Europe.
European Commission Appendix A
DG Internal Market
 98
APPENDIX A: Methodology
1. Information sources
The data sources we used in the context of our analysis are the following:
- EFAMA;
- LIPPER;
- The funds' financial statements;
- Fitzrovia;
- Other applicable surveys and statistics.
- EFAMA is the representative association for the European investment management
industry. It provides a relevant source of information for data on the European
Investment Management market and describes its main components. EFAMA gives an
idea of the market’s reality to identify the most active / representative countries in the
asset management industry in Europe. We have used EFAMA statistics as a benchmark
to help us build our sample and ensure that it is representative in terms of market
composition. Details on data used in this context are described below.
- LIPPER, a Reuters company, is a global leader in supplying investment fund
information, analytical tools and databases. The databases accounts for nearly 80% of
the funds in Europe and provides detailed information on every single fund. It is
considered an official source of investment management information and recognized as
an assertive data provider. Whereas EFAMA gives us some global information on the
market and its components, data included in the Lipper database enabled us to identify
and select the funds to be included in our sample based on defined criteria.
We used both databases developed by Lipper for UCITS and non-harmonised funds. In
a first instance, the information that was considered was the legal form and
domiciliation country of the selected funds. Within each of these categories, we used the
asset type, total net assets, and launch date of the funds as key selection criteria. The
latter set of data was captured not at fund (legal entity) level but at sub-fund level, in
order to obtain more specific information.
European Commission Appendix A
DG Internal Market
 99
- We also reviewed the financial statements of the selected funds, which are the basic
financial information available to shareholders and make it possible to report the main
funds’ net asset components accurately. Any information presented in our report
concerning the portfolio composition and the embedded leverage of the selected funds
(where applicable and available) is based on disclosures made in the financial
statements. We collected the financial statements either through the promoters’ website,
dedicated national websites (such as CCLUX in Luxembourg for instance) or through
direct contact with promoters. We never used our status as statutory auditor of the
selected funds (where applicable) to collect the financial statements directly through our
local offices in the surveyed countries. PwC worldwide is not allowed to provide third
parties with financial information concerning its clients, as it is subject to professional
secrecy. This means that local offices treat PwC Luxembourg as a third party and that
any financial information they may have on the funds is privileged. When no financial
statements were available, we also used quarterly fact sheet or other marketing
documents produced by fund managers and giving some information on the NAV and
portfolio composition.
- FITZROVIA is a data provider that complements the information provided by EFAMA,
especially concerning the analysis of Luxembourg and Irish markets.
- Other market surveys or statistics provided by recognized professional/legal authorities
have also been used, especially in the context of non-harmonised funds where financial
statements were more difficult to obtain due to confidentiality reasons.
2. Key data retrieved from Lipper database
As indicated above, the one of the data sources we used is Lipper.
Performance analysis mostly relied on information collected in the Lipper database, which
represents a large universe of funds in existence. Furthermore, Lipper includes a
sub-database dedicated to non-harmonized funds (including hedge funds) that has around
7,000 funds registered worldwide.
We selected the Lipper database, as it contains price information on a daily basis (net of
fees charged directly to the fund, but not of fees charged to investor when subscribing or
redeeming) as well as additional information, including descriptive statistics and
qualitative data, such as legal entities and sub-funds. As already indicated, in our
methodology, the term “funds” refers to “sub-funds” in case of umbrella structure.
Along with prices, other key data gathered from this source are NAV date, ex-dividend
date, asset type, launch date and domiciliation country of the selected funds.
- NAV Price
The NAV price used is the bid or offer price at the end of each dealing day converted into
EUR currency. The price has been used along with ex-dividend date in order to cancel the
impact of the dividend distribution in the fund performance calculation.
The data covers the period starting January 1
st
, 2002 through December 31
st
, 2006.
European Commission Appendix A
DG Internal Market
 100
- Asset type
The funds are classified by Lipper into investment styles on the basis of:
- Investment objective (statement of the promoter)
- Prospectus
- Fund Fact Sheet of promoter group
The investment objective as stated by the promoter and the prospectus are the primary
sources for the classification made by Lipper. Each fund is assigned one of the following
asset types according to the assets in which it invests. Funds with a strategic diversification
in equity and bond securities along with ancillary liquid assets are considered Mixed Asset
funds. Where a strategic use of derivative/alternative instruments does not allow an asset
type to be assigned, the fund is placed in the “Other” category.
Asset Type Primary Investment Objective
Equity Focused on equity market investments
Bond Investments in fixed income instruments with
an average maturity longer than 1 year
Money Market Investments in fixed income instruments with
an average residual life to maturity of less than
12 months
Mixed Assets Strategic mix of equities and bonds
Other Capital guaranteed, Protected, etc.
This classification is made at the time the fund is entered in the database and is not
adjusted in order to reflect any change in actual portfolio composition.
For the purpose of the analysis, the classification provided by Lipper has been used for the
aggregate analysis of funds under UCITS III legislation. Conversely, for the
non-harmonized sample, we created a new set of categories, which includes:
- Hedge funds;
- Funds of hedge funds;
- Real Estate funds;
- Private Equity funds; and
- Other forms of investments.
- Launch Date
The launch date corresponds to the date the fund is launched along with its initial price.
This information allows us to separate funds launched before and after the UCITS III
directives were implemented thus enabling us to compare the results from the two sub-
samples.
- Domicile
The information has been used to retrace the fund based on its country of domicile.
European Commission Appendix A
DG Internal Market
 101
3 Sampling methodology
One of the main characteristics of the observed fund population is the high degree of
concentration of assets under management within a small number of funds. A short
analysis of the detailed European fund population, based on financial information available
in the Lipper database and described below, shows that more than 80% of assets belong to
less than 20% of funds. This characteristic is observed whatever the domicile or type of
investment fund.
Additionally, the funds are not equally distributed within European countries and more
than 50% of the funds are domiciled in Luxembourg and France (EFAMA statistics
2005). The remaining funds are then mostly concentrated in a small number of countries.
Moreover, the analysis of fund type by country, by type of investment and by legal status
(UCITS vs. non-harmonised funds), shows that the distribution structure in each country is
different, as evidenced in the graphs below:
European Commission Appendix A
DG Internal Market
 102
Investment type - UCITS
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
France
Germany
Ireland
Italy
Luxembourg
Poland
Slovakia
Spain
UK
Bond Equity Mixed Assets Money Market Other
Source: Lipper Hindsight, end October 2006
As these tables show, the European fund population is not homogeneous enough to meet
the requirements needed to apply a classical sampling technique: applying a random
technique could cause us to miss a needed population, especially in the case of large funds
or funds of small countries. Consequently, the sampling methodology had to be tailored to
capture the specific characteristics of the funds’ population in each country. The latter is
the methodological approach we have adopted and which is guided by factual, statistical
data and also by our knowledge of the fund market in Europe and our understanding of its
key features and characteristics.
Investment type - Non-UCITS
0% 20% 40% 60% 80% 100%
France
Germany
Ireland
Italy
Luxembourg
Poland
Slovakia
Spain
UK
Bond Equity Mixed Assets Money Market Other
European Commission Appendix A
DG Internal Market
 103
To summarize, our sampling approach responds to the following key objectives:
- Accurately represent the European investment management industry,
- Respond to the requests and results expected by the European Commission.
- Methodology applied for UCITS
Once the proportions were defined - both in terms of country and asset class based on latest
EFAMA statistics - we had to select the funds to be included in our sample. In order to
cover a representative proportion of investors, we added one criterion relating to the size of
the funds and one constraint relating to the minimum coverage of the respective countries
to be included in our sample.
Therefore, we set the coverage of our sample to a minimum of 20% of the assets under
management in each local market. Within this population, we ensured that we captured the
largest funds within each country and asset class, which are supposed to include a large
number of investors. In order to avoid any bias in our sample resulting from a selection
limited to largest funds, some smaller funds were selected on a random basis within the
remaining population. This methodology enabled us to consider whether the size of a fund
may have impacted the use of the wider investment powers of UCITS funds, and whether
the size may have had an impact on fund performance and volatility.
If investment strategy and fund size were considered key parameters for building our
sample, funds also had to be broken down according to their launch date. This factor was
instrumental in identifying whether the funds launched directly under the UCITS III
regulation behaved differently to the funds launched under the previous regulation. To
compare this, our sample was split into funds launched under the older directive (UCITS I
6
i.e. before December 31, 2002), and funds launched directly under the new legal
framework (UCITS III, i.e. after December 31, 2002).
The analysis of fund behaviour during the grand-fathering period itself and in the light of
its actual conversion date from UCITS I to UCITS III is not a key element in our analysis,
as the exact conversion date of the respective funds in our sample is not a piece of
information that can be easily captured. This particular period will be further analysed
through face-to face interviews conducted with promoters.
The resulting sample is very close to the picture described by EFAMA for the year ended
December 31, 2005 (latest statistics available when we defined our population) but
deviates slightly for some asset classes and / or countries, as we had to meet the additional
criteria listed above. The resulting sample is described in section 3.2.
6
The term “UCITS I funds” refers to funds falling under Directive 1985/811/EEC, applicable to funds before
the introduction of UCITS III.
European Commission Appendix A
DG Internal Market
 104
 Methodology applied to non-harmonised funds
The selection of the non-harmonised fund sample could not exactly follow the same rules
as UCITS funds because, by definition, this market is not harmonised and the legal
framework applicable to this kind of funds is specific to each country. In addition, the
maturity of the respective markets is extremely different from one country to another in
terms of both market size and investment vehicles / strategies available to investors. In
practice, we noted the predominance of particular investment types in some countries as
well as the inexistence of some legal structures in some others.
EFAMA statistics describe the investment fund population and the proportion between
UCITS and non-harmonised funds. In addition, they also indicate the weight of the various
investment strategies used by non-harmonised funds (hedge funds, funds of hedge funds,
real estate funds, etc). However, the definition of non-harmonised products as per EFAMA
may differ from ours as defined above; we have not always identified statistics detailing
the structure of each local market. In that context, we used the market information that was
available as a starting point together with our extensive knowledge of the European
investment management market. Where necessary, we also relied on the support of our
local PwC offices in each of the countries surveyed. In addition, we looked for
complementary information from the following sources: the AIMA, (“Alternative
Investment Management Association”), the INREV, (“European Association for Investors
in Non-listed Real Estate Vehicles”) and the EVCA, (“European Private Equity and
Venture Capital Association”).
The countries in our scope are the same as for UCITS funds. As already indicated above,
our objective is to cover all types of investment strategies and vehicles available. The only
constraint is that selected funds have a minimum one year-record (since launch date), so
that we are able to evaluate fund performance and volatility over a representative period of
time.
European Commission Appendix B
DG Internal Market
 105
APPENDIX B: Methodology of performance/risk analysis
This appendix describes the formulas / procedures supporting the results presented in
section 3 of this report, together with the key data used.
1. Return Calculation
a) Basic returns
The method for calculating the historical return on a fund for a given period is the relative
variation of the NAV price of the fund over the period, adjusted, if applicable, by the
dividend payment. The NAV information used is net of fees born and paid directly out of
the Fund net assets ; these encompass mainly administrative and management fees and
does not integrate any subscription / redemption fee paid by the shareholder as a
remuneration of the intermediary / distributor. The return is given by:
1
1
÷
÷
+ ÷
=
it
it it it
it
NAV
D NAV NAV
R
Where
it
NAV
= is the NAV price of the fund i at the time t;
1 ÷ it
NAV
= is the NAV price of the fund i at the time t-1;
it
D
= is the dividend of the fund i paid at time t.
There is however one event we have to identify in the respective NAV’s data, which could
lead to a bias in our computation: dividend distribution. On the day when dividend
payment is booked (ex-date), the NAV per share of any distributing funds decreases by the
amount corresponding to the dividend payment. As distribution policy differs depending
on the investment strategy and domiciliation country of the respective funds, we had to
cancel the impact of this transaction on our performance computation.
In order to perform this, one option would be to reintegrate on each ex-dividend date the
amount distributed into the NAV per share information we have. This implies however that
we need to know on each single dividend date the dividend paid and convert it into EURO
to reintegrate this in our NAV data. This results in a huge work, which can be source of
error. Alternatively, we decided to exclude the return of the fund on the ex-dividend date
(i.e. the date the share price drops by the amount of dividend) from the calculation, thus
removing impact on average performance. All ex-dividend dates of the selected funds and
share classes have been retrieved from Lipper and exclusion has been made on this basis.
Hence, the approach takes the following formula as a historical return on a fund:
1
1 ,
,
,
÷ =
÷ t i
t i
t i
NAV
NAV
R
European Commission Appendix B
DG Internal Market
 106
We thus compared the value of the fund at the end of the period with its value at the
beginning of the period. This calculation provides a rate of return that depends on the
evolution of the NAV over the period. In the context of our analysis, the period considered
is one day, meaning we computed daily returns.
b) Return Data Frequency
It has become a common practice for the investment management industry to carry out a
daily observation periodicity.
We therefore constructed our analysis using observed daily prices. However, it should be
stressed that this can have a negative impact since it also influences the distribution of
observed excess returns. In fact, the assumptions guiding assets returns (returns are
normally distributed, there is no serial correlation between investment returns) are not
always verified for daily return data.
Finally, we had to cope with some specific features regarding the daily return time series.
The first difficulty deals with the existence of outliers that are observation with extremely
large value and very low frequencies (usually due to an incorrect valuation of the fund).
Specifically, the distribution was trimmed to exclude abnormal values as these could have
led to misleading inference about the average performance.
The second difficulty refers to the case of infrequent price evaluations. In some series, we
observed that a large performance on one day is usually preceded by periods with
stationary prices, which could be interpreted as possible evidence that the fund is
infrequently evaluated, for example once a week, or adjusted occasionally. We then had to
recalculate the return series on a daily basis using the following formula:
( ) 1 1 *
1
, ,
÷ + = n
t i t i
R R
Where
n is the number of days without price changes;
t i
R
,
is the observed performance after a period of no revaluation; and
t i R ,
*
* is the daily recalculated return during the stationary period.
We applied this methodology, even if we are aware that this may slightly lower the
volatility of some funds with less frequent price observations as price evolution between
2 price observations is somehow linearized. However, the impact of this is not significant
as, in our computations; we only included funds, which had sufficient number of
observations. In addition, we preferred this bias to a situation where money market funds
or other funds with low daily variations or small NAV per share, which result in a nil
variation over a certain period of time, followed by a more important variation and results
in a high volatility, which is not representative of the risk in price change associated with
this type of funds.
European Commission Appendix B
DG Internal Market
 107
c) Autocorrelation in return series
The presence of illiquid assets in the portfolio may generate autocorrelation since prices of
less active assets appear not to incorporate some of the recent information that is already
contained in the prices of more liquid assets.
Generally, positive autocorrelation of returns is present if a positive (negative) return in
one period is followed by another positive (negative) return the next. Negative
autocorrelation is present if positive (negative) returns are followed by negative (positive)
returns.
The problem with a historical series with positive autocorrelation is that this will have a
lower volatility than an uncorrelated series. As less volatility is usually preferred to more,
the attractiveness of returns is therefore distorted (overestimated) in the presence of
positive autocorrelation. This can lead to an overestimate of the Sharpe ratio for example
so that any decision based on this indicator can be dramatically wrong.
In order to identify the presence of autocorrelation on historical return series, we
performed the Ljung-Box test, which is based on the Q-statistic.
( )
(
¸
(

¸

÷
+ =
¿
=
m
k
k
k n
r
n n Q
1
2
) (
) 2 (
where
n is the sample size;
m, is the number of autocorrelation lags included in the statistic; and
k r
2
is the squared sample autocorrelation at lag k (k=1 in our context).
We test the null hypothesis (absence of autocorrelation in the series of returns) against the
alternative hypothesis (the series of returns are autocorrelated):
H0: No presence of autocorrelation
H1: Series autocorrelated
The Q statistic follows a Chi-2 distribution with 1 degree of freedom. If the P-value
associated to the Q statistic is less than 5%, we reject the null hypothesis.
Historically, positive autocorrelation has been observed in real estate and hedge fund time
series. If this situation can somehow lower the volatility of this type of funds one year, it
has no impact on the analysis of the evolution of returns over the 5 year period.
European Commission Appendix B
DG Internal Market
 108
Once we identify the presence of autocorrelation, the approach we used to remove the
impact of autocorrelation has been to “unsmooth” the observed returns to create a set of
new returns. Following the Geltner method (1991), it is possible to derive an expression
that will yield an unsmoothed series of returns with zero first order autocorrelation:
i
t i i t i
t i
r
R r R
R
÷
× ÷
=
÷
1
1 , ,
,
*
Where
t i R ,
*
is the unsmoothed return;
t i
R
,
is the real return at time t; and
i
r equals to the autocorrelation coefficient at lag 1.
The newly constructed series will have the same mean as
t i
R
,
and zero first order
autocorrelation.
2) Statistics re-weighting
The process described below has been applied only on the UCITS sample and population.
Conversely, no such calculation has been done for the global fund population.
As we considered EFAMA statistics as the most representative of the European Fund
industry that provide a realistic picture of the European Fund population, we designed our
sample on its basis, even though we extracted our sample from the fund population
available in the Lipper database.
However, we had to slightly derive from this standard in order to cover, for the respective
asset classes and countries, a minimum of 20% of the assets under management.
In order to eliminate the bias resulting from the difference in weighting in our sample and
in the Lipper database, compared to the European population described by EFAMA, we
revised our calculations and weighted both the sample and population statistics on the basis
of EFAMA’s proportions.
We compute the new weight of each country (c) for each asset type (t) such as:
t c
t c
t c
p
P
W
,
,
,
=
Where P is the proportion in the EFAMA population and p is the proportion in the Lipper-
Reuters database or in the sample.
Calculation of weighted means of statistics (S) for each asset type (t) is therefore:
¿
¿
=
=
=
1
,
1
, , ,
i
n
t c
i
n
t c i t c
t
W
S W
M
European Commission Appendix B
DG Internal Market
 109
3) Summary Statistics
We ran our analysis of returns over the entire sample period (2002 – 2006) as well as for
each year separately. We provided summary statistics for all previously defined asset
classes within the sample representing the simple average of each statistic. The summary
statistics include mean, standard deviation, median, skewness (Skew), kurtosis, minimum
(Min) and Maximum (Max).
Arithmetic Mean
We computed the arithmetic mean of returns for the sub-periods using the following
formula:
¿
=
=
N
t
t i i R
N
R
1
,
*
* 1
where N is the number of periods.
We used the arithmetic mean of returns from past periods since it has the following
advantages: provides an unbiased estimate of return for the period; easily understood by
the public; does not provide results that are significantly different from those of a
geometric mean where the main difference is that the arithmetic mean does not assume the
reinvestment of generated income.
Standard deviation
Following the modern portfolio theory developed by Markowitz, the concept of average
return is not sufficient to evaluate the attractiveness of an investment and its distribution.
We calculate variance and standard deviation as a measure of dispersion of a fund’s return.
The dispersion is defined as the variability around the central tendency (i.e. the mean of the
distribution). In finance and investment theory, the central tendency is the measure of the
reward and the dispersion is a measure of risk. The statistical measurements are the
variance σ
i
2
and its square root, the standard deviation σ
i
.
When we evaluate a sample of n observations, the variance is calculated using the
following formula:
( )
2
1
2
1
¿
=
÷ =
N
t
i it i
R R
N
o
Unlike the mean, the computed variance is squared units of measurement. The problem of
interpreting is solved by using the standard deviation, the square root of the variance,
calculated as follows.
( )
2
1
1
¿
=
÷ =
N
t
i it
R R
N
o
The variance (or better still, the standard deviation) is the most widely used measure of
risk. The disadvantage of this measure is that it considers the risk of above average returns
and the risks of below average returns in the same way, while an investor is only worried
about below average returns. Further, in a case where returns are not symmetric (i.e.
skewness) the variance becomes less reliable.
European Commission Appendix B
DG Internal Market
 110
Normality of return series
In investment performance analysis, a lot of measures require the assumption of normally
distributed returns.
The Normal distribution N (μ, σ2) has the following key properties:
 It is completely described by its first two moments (i.e. mean and variance);
 The normal distribution is unimodal (one value appears most frequently);
 Skewness = 0, meaning that the distribution is symmetric about its mean, and Kurtosis
= 3;
 A linear combination of normally distributed variables is also normally distributed.
In order to take into account the fact that some return series may not be normally
distributed, we studied certain descriptive statistics (e.g. skewness and kurtosis). It is
important to notice that the skewness and the kurtosis of the distribution may vary with the
horizon used to calculate the returns since moments of the distribution change over time.
As described below, however, they can provide us with more information on return series
that are not normally distributed.
Median
The median is an ordered statistic which is defined as the 50th percentile of the return
distribution, that is, half the observation lies above the median and half below. The median
is important because the arithmetic mean can be affected by extremely large or small
values. When this occurs, the median is a better measure of central tendency than the mean
as it is not affected by extreme values.
The median is calculated after arranging data from the highest to the lowest value and
finding the middle observation.
This measure is not used regularly in performance analysis but we decide to include it in
ours since the distribution of fund returns may not be normal. In fact, as an ordered
statistic, the median is a more robust measure of the tendency of price changes that make
up distribution if the latter is not normal.
Skewness
Distributional symmetry implies that loss and gain intervals will exhibit the same
frequency. Skewness refers to the extent to which a return distribution is not symmetrical.
It is important because the degree of symmetry (or skewness) tells us if deviations from the
mean are more likely to be positive or negative. Non symmetrical distribution may be
either positively or negatively skewed.
Positively skewed distribution is characterised by more likely but smaller losses and less
likely but extreme gains. In other words, negative events are more frequent but limited in
magnitude. Negatively skewed distribution is conversely characterised by more likely but
smaller gains and less likely but extreme losses, this means that bad events are more likely
to be extreme even though they occur less frequently.
European Commission Appendix B
DG Internal Market
 111
Skewness affects the location of mean, median and mode: For instance, a symmetrical
distribution, the mean, median and mode are equal and the skewness is zero.
For a positively skewed distribution, the mode is less than the median, which is less than
the mean.
Positive skewness = Mode < Median < Arithmetic Mean,
Negative Skewness = Mode > Median > Arithmetic Mean,
Symmetrical = Arithmetic Mean = Median
Kurtosis
The Kurtosis measures both “peakedness” and tail heaviness of a distribution relative to
that of normal distribution.
A leptokurtic distribution is more peaked than normal distribution meaning that it will have
a greater percentage of small deviations from the mean and a greater percentage of
extremely large deviations from the mean (fatter tails). With regard to an investment
returns distribution, the greater the likelihood of a large deviation from the return, the
greater the perception of risk.
The computed kurtosis for a normal distribution is three and Excess Kurtosis is defined as
kurtosis minus three. A leptokurtic distribution has, therefore, excess kurtosis greater than
zero.
Consequently, a kurtosis 3 generally indicates some potential risks regarding negative
events or possible losses. It may be a useful indicator which supplements the risk resulting
from a Value at Risk calculation based on a parametric approach.
4) Performance and risk
a) Return Analysis
In analysing fund performances for both UCITS and non-harmonised funds, we firstly
considered the average of returns taken from the descriptive statistics of the sample. All the
data was adjusted to yield a zero first order autocorrelation in the return series. It is
important to notice that the performance as well as risk estimates vary according to the
base currency (i.e. EUR) of the prices used to construct the return series. Furthermore, it is
common investment knowledge that past performance is not a reliable indicator of future
performance.
Implication of Survivorship Bias
As with the evaluation of historical data for any investment instrument, certain biases may
exist. The most significant bias in the hedge fund industry for example, is represented by
the survivorship bias, that is, the fact that the database only includes ongoing funds. Funds
may drop off the database due to various reasons such as mergers and acquisitions, closure
or liquidation, and voluntary withdrawal. Poor performance however, is generally a major
reason.
European Commission Appendix B
DG Internal Market
 112
The effect of the survivorship bias is greater for hedge fund data than for other asset
classes because of the lack of required reporting standards in the industry. Fund managers
tend to “cherry pick” the information they choose to release, reporting on their more
successful funds while not providing information on poorly performing or defunct funds.
Reported returns for a hedge fund database are, therefore, overstating performance because
of survivorship bias. On the other hand, hedge funds with highly volatile returns tend to
fail more frequently, and defunct funds are generally not included in a database. For this
reason, the risk measure of hedge funds as an asset class would be understated.
However, the results for the hedge funds and the funds of hedge funds included in our
sample are less subject to the effect survivorship bias for the following reasons:
First, they are all domiciled in Europe and are therefore subject to the supervision of a
member state of the European Union.
Second, they are all created under clear regulatory frameworks, which include strict
reporting requirements.
Finally, since they are all domiciled in Europe, they use third party service providers for
the NAV calculations. In the US, it is more common to find hedge fund managers
calculating their own NAV, which may raise concerns regarding independence and
governance.
Performance
Our sample statistics are based on the arithmetic mean of return. However, when
evaluating returns over long period (especially one year), it is always preferable to
compute the geometric mean, that is, the real growth of the investment over the whole
period as it assumes that all the intermediate incomes are reinvested. The rate of return is
given by the formula:
( ) 1 1
1
1
,
÷
(
¸
(

¸

+ =
[
=
T
T
t
t i G
R R
Where
N is the number of observations; and
it
R
is the return of the fund i for period t.
In general, the results from geometric mean are close to those of arithmetic mean.
However, geometric mean is always less than the arithmetic mean, unless the
it
R
return is
equal in which case the two means are identical and the differences increase as the
dispersion of observations increases from period to period.
European Commission Appendix B
DG Internal Market
 113
Annualized return
We present the effective annual performance using an annualized value based on a 260-day
year of the computed geometric mean. It is the annual return that would result in the same
return if compounded annually, using the following formula:
Annualized returns have the important advantage of taking into account the reinvestment
of dividends and capital gains as well as the price changes (NAV) of the fund over a
specific period of time. It is a hypothetical rate of return that, if achieved annually, would
have produced the same cumulative return if performance had been constant over the entire
period.
The result is a very important tool in an investor's decision-making process. A fund that
has been able to achieve a consistent return over longer periods of time is considered to
have more value to an investor's planning. The mark of a good fund is, therefore, one that
has produced very consistent annualized returns over varying periods of time.
Removing market effect
We used a model taken from the Capital Asset Pricing Model to remove the market impact
on performance figures. This is, however, more specifically related to equity portfolios.
This method evaluates fund returns in relation to the market return and the sensitivity to
the market.
In particular, alpha or Jensen’s alpha is defined as the differential between the return on the
portfolio in excess of the risk-free rate and the return explained by the market model. This
is calculated by performing the following regression:
The
i
o
measures the proportion of excess return that is due to the manager’s choice. The
market is represented by published style indices that correspond to the most appropriate
benchmark for each fund based on its investment style and geographical allocation. This
measure is still not risk adjusted as it does not allow portfolios with different levels of risk
to be compared: In fact, the level of alpha is proportional to the level of risk taken, as
measured by the beta.
In order to evaluate the goodness of fit of our model, we evaluate the R2 of the regression.
The main advantage of this approach is that it will directly link the fund’s returns to
explicit observable factors (i.e. market indices) to derive relative performance measures
that are not affected by the sensitivity to market conditions.
However, we should mention the fact that observable market risk factors are included in
the model through a discretionary choice. The model could then display poor explanatory
powers (as reflected by a low R2).
Furthermore, due to the wide range of instruments and techniques available to fund
managers, the model presents the risk of overestimating the alpha. This is particularly true
for fund returns that can show non-linear exposures to standard asset classes.
( ) 1 1
260
÷ + = G R return annualized
t i t free BMK i i t free it
R R R R
, , ,
) ( c | o + ÷ + = ÷
European Commission Appendix B
DG Internal Market
 114
b) Risk Analysis
Calculation of risk measures for portfolios of assets, funds for instance, may involve a
position level analysis, i.e. analysis of the risk of each asset holding, that consists in a
multivariate model setting.
However, the complexity of dealing with the large number of information for the funds in
the sample, the difficulty to disaggregate the risk of large and complex positions, has lead
the analysis toward a univariate approach.
We ran our analysis firstly with the simplest measure of risk, based on the Markowitz
portfolio theory: the variance or the standard deviation of returns. We then provide more
sophisticated downside risk measures that are now extensively used in investment practice:
Value at risk.
Annualised Volatility
We define volatility as annualized standard deviation. The standard deviation of the return
time series is the one calculated for the descriptive statistics of the sample.
The standard deviation represents historical volatility calculated from daily data. This
means it is a daily volatility. Because volatilities are usually quoted on an annual basis,
daily historical volatilities are generally converted to an annual basis by applying the
square root of time rule.
Following the square root of the time rules, volatility increases with the square root of the
unit of time, based on a 260 days year, as follows:
260 × =
daily
annualised o o
The resulting volatilities are referred to as annualized volatilities - as opposed to annual
volatilities - to the fact that this is just a quoting convention.
Value at Risk
As we already discussed, standard deviation considers both the risks of above average
returns and those of below average returns.
A downside risk measure for traditional investment portfolios is the Value-at-Risk (VaR),
that is the worst loss that can happen under normal market conditions over a specified
horizon (e.g. 1 day) at a specified confidence level (e.g. 95% or 99%). VaR has been an
established risk measurement and reporting tool in the banking industry for several years
and has become part of the hedge fund risk managers’ risk analysis. UCITS III
sophisticated funds are extensively using Value at Risk to monitor the global exposure on
derivatives. This is further explained in section 6 relating to the comparative regulatory
analysis.
It can be seen in this section that most sampled countries’ regulators have implemented
value at risk with a confidence level of 99% even if no method/approach is imposed.
European Commission Appendix B
DG Internal Market
 115
We compute VaR through the model-free (or unconditional mean/variance) method (i.e.
historical simulation), as well as through mean/variance methods (e.g. VaR Gaussian).
Among these models, technical specifications can address autocorrelation, asymmetry and
fat tails.
Historical VaR:
The simplest way to estimate VaR is to use the sample quantile estimate based on
historical return data. The method is an unconditional model approach that has no
assumptions about the distribution and is referred to as Historical VaR.
Since Historical VaR is based on the unconditional distribution of losses, it allows
departures from normality to be captured to some extent. As previously mentioned,
however, historical returns may be affected by autocorrelation. It is thus necessary to
adjust returns to remove any serial correlation in return series.
Parametric VaR:
Alternatively we can focus on another method (parametric or conditional models) that uses
the same assumption about the return distribution: Gaussian VaR.
As proposed by Riskmetrics
TM
, the traditional approach to estimate parametric VaR
assumes that returns follow a normal distribution, completely characterized by its mean
and standard deviation. The normal VaR is thus calculated by the following formula:
o o µ
o
× + =
÷
) (
1 , 1
z VaR
Where
µ
and
o
are, respectively, the sample mean and standard deviation of returns; and
) (o z
is the confidence level multiplier to arrive at the tail probability of loss levels and
implied VaR confidence levels (Confidence levels are 1- α).
The main disadvantage of this method is that it strongly relies on the assumption of normal
return distribution. For this reason the results produced may be different from those
obtained from historical simulation, especially in presence of asymmetry or fat tails.
Comparing results based on both approaches is of particular interest.
c) Performance and Risk-adjusted measures
Following portfolio theory, we quantitatively measure the link that exists between risk and
return by producing a risk adjusted measure widely used by investment firms, the reward
to variability ratio or Sharpe Ratio.
Sharpe Ratio
The ratio measures the return of a fund in excess of the risk-free rate, compared to the total
risk of the fund, measured by its standard deviation.
i
free i
i
R R E
S
o
÷
=
) (
European Commission Appendix B
DG Internal Market
 116
Where
free
R
is the risk free rate that is represented by a money market rate (Euribor 6m), and
) (
i
R E
and
i
o
are the average rate or return and the standard deviation of the fund
respectively.
This measure is based on the total risk (i.e. standard deviation) so it enables the relative
performance of funds to be evaluated and for this reason is widely used by the investment
industry.
Whilst its extensive use, the ratio assumes that fund performance can be reduced to the first
and second moments of the return distribution (its mean and variance).
When fund returns are not normally distributed however, the mean and standard deviation
are not sufficient to describe the distribution. The Sharpe ratio suffers, therefore, from the
same problem described for the normality assumption. Consequently, for those funds, the
Sharpe ratio is not enough to appreciate the link between risk and return. Other statistics
(historical Value at Risk, skewness and kurtosis) enable us to supplement the analysis.
European Commission Appendix C
DG Internal Market
 117
APPENDIX C: Geographical focus of UCITS equity funds.
2002
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
ly
L
u
x
e
m
b
o
u
r
g
P
o
l
a
n
d
S
l
o
v
a
k
i
a
S
p
a
i
n
U
K
Undefined
UK
Spain
Poland
Other European countries
Other
Luxembourg
Italy
Ireland
Germany
France
Emerging Markets
Asia
America
2003
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
ly
L
u
x
e
m
b
o
u
r
g
P
o
l
a
n
d
S
l
o
v
a
k
i
a
S
p
a
i
n
U
K
Undefined
UK
Spain
Poland
Other European countries
Other
Luxembourg
Italy
Ireland
Germany
France
Emerging Markets
Asia
America
European Commission Appendix C
DG Internal Market
 118
2004
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
ly
L
u
x
e
m
b
o
u
r
g
P
o
l
a
n
d
S
l
o
v
a
k
i
a
S
p
a
i
n
U
K
Undefined
UK
Spain
Poland
Other European countries
Other
Luxembourg
Italy
Ireland
Germany
France
Emerging Markets
Asia
America
2005
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
ly
L
u
x
e
m
b
o
u
r
g
P
o
l
a
n
d
S
l
o
v
a
k
i
a
S
p
a
i
n
U
K
Undefined
UK
Spain
Slovakia
Poland
Other European countries
Other
Luxembourg
Italy
Ireland
Germany
France
Emerging Markets
Asia
America
European Commission Appendix C
DG Internal Market
 119
2006
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
ly
L
u
x
e
m
b
o
u
r
g
P
o
l
a
n
d
S
l
o
v
a
k
i
a
S
p
a
i
n
U
K
Undefined
UK
Spain
Poland
Other European countries
Other
Luxembourg
Italy
Ireland
Germany
France
Emerging Markets
Asia
America
European Commission APPENDIX D
DG Internal Market
 120
APPENDIX D: Detailed Allocation of Bond and Money Market Funds.
2002
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
ly
L
u
x
e
m
b
o
u
r
g
P
o
l
a
n
d
S
p
a
i
n
U
K
Undefined
High yield + emerging
Developed countries
2003
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
ly
L
u
x
e
m
b
o
u
r
g
P
o
l
a
n
d
S
l
o
v
a
k
i
a
S
p
a
i
n
U
K
Undefined
High yield + emerging
Developed countries
European Commission APPENDIX D
DG Internal Market
 121
2004
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
ly
L
u
x
e
m
b
o
u
r
g
P
o
l
a
n
d
S
l
o
v
a
k
i
a
S
p
a
i
n
U
K
Undefined
High yield + emerging
Developed countries
2005
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
ly
L
u
x
e
m
b
o
u
r
g
P
o
l
a
n
d
S
l
o
v
a
k
i
a
S
p
a
i
n
U
K
Undefined
High yield + emerging
Developed countries
European Commission APPENDIX D
DG Internal Market
 122
2006
0%
20%
40%
60%
80%
100%
F
r
a
n
c
e
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
ly
L
u
x
e
m
b
o
u
r
g
P
o
l
a
n
d
S
p
a
i
n
U
K
Undefined
High yield + emerging
Developed countries
European Commission APPENDIX E
DG Internal Market
 123
APPENDIX E: CSSF statistical reporting
This appendix provides with additional information collected in the CSSF activity reports,
relating to NAV calculation error or non-compliance with investment restrictions reported
to the supervisory authority.
The evolution from 2002 to 2006 has been the following, concerning increase from one
year to another of the number of NAV calculation errors or non compliance with
investment restrictions (“investment breach”) reported. In order to better assess this
evolution, we inserted also the evolution of assets under management of funds under part I
and part II of the law of December 20, 2002, excluding therefore funds dedicated to
institutional investors.
In terms of quantities, in 2006 1150 incidents were declared to the CSSF, 274 relating to
NAV calculation errors and 876 resulting from non-compliance with investment
restrictions.
Concerning NAV calculation errors and their financial implications, around 67% of the
errors resulted in compensations inferior to EUR 25 000 for the concerned fund and
EUR 2 500 for individual shareholder (if they left the Fund in the meantime or if they were
impacted when subscribing into one Fund), whereas 32% resulted in higher compensation
that the amount indicated above, to the fund or to individual shareholders. It has to be
noted that in 2003, the repartition was 54% for resulting compensation below the threshold
defined before, and 46% above and the statistics concerning the years in the meantime
remain within the 2003 and 2006 levels.
40.00
60.00
80.00
100.00
120.00
140.00
160.00
180.00
200.00
220.00
240.00
0 2003 2004 2005 2006
NAV error
Investment breach
Assets under management
European Commission APPENDIX E
DG Internal Market
 124
Concerning the compensations made following correction of non-compliance with fund
investment restrictions, 65% of these situations, did not result in any compensation as
neither the fund, nor investors did suffer from this situation. For the remaining part, 27%
resulted in compensation below EUR 25 000 for the Fund and EUR 2 500 for a single
shareholder whereas 8% resulted in higher compensations. In 2003, the statistics were 50%
of cases did not result in any compensation, 26% with payment below EUR 25 000 and
EUR 2 500 respectively, the remaining 24% ended up with indemnification over
EUR 25 000 for the fund or EUR 2 500 for any single shareholder.
The graph and statistics enclosed above tend to show that the number of NAV computation
errors or investment breach in Luxembourg funds has increased since 2004, but less
rapidly than the size of assets under management. In addition, the percentage of incidents
that required compensation to the fund or to the shareholders has decreased during the
same time frame as it went from 50% in 2003 down to 35% in 2006.
When considering the origin of the NAV calculation errors, in 2006 the repartition is the
following:
Origin of errors -2006
Pricing errors
18%
Acccounting entries
43%
Expenses and
accruals
15%
Trade processing
10%
Swap and futures
valuation
7%
Others
7%
European Commission APPENDIX E
DG Internal Market
 125
In 2002, the situation was the following:
In the light of these statistics, we can identify that despite the wider use of derivatives and
presence of more sophisticated products, the weight of pricing errors and errors generated
by Swaps and futures did not increase significantly.
The main areas of growth are accounting entries and trade processing, which may result
from the difficulty to account new products and to integrate them in the daily business.
However no additional information is provided concerning these specific points by the
CSSF and other elements may also need to be considered.
If the risk profile of UCITS has slightly changed over the year, some investments remain
dedicated to the non-harmonised environment as reported in section 5.1.
Origin of errors -2003
Pricing errors
29%
Acccounting entries
32%
Expenses and
accruals
17%
Trade processing
4%
Swap and futures
valuation
6%
Others
12%
European Commission APPENDIX F
DG Internal Market
 126
APPENDIX F: Importance of operational risk.
Our analysis can be complemented by results of interviews conducted in 2003 by Oxera
(2006)
7
where European Asset Managers were asked to identify the most important
operational risks the industry faces and to evaluate their frequency and their potential
impact. The results of these interviews are summarized as follows:
7
Oxera, “Current trends in asset management”, 2006
European Commission APPENDIX G
DG Internal Market
 127
APPENDIX G: Survey on Fund failure.
Origin of actual fund failure can give additional information on the importance and
potential implications of main risks funds are exposed to. In this context, the analysis
presented in this appendix can bring additional background to this question.
Operational issues are in fact a main risk factor to explain that on average approximately
“15 funds collapse per year (to be compared to approximately 350 hedge fund closures per
year) out of a universe of a few thousand funds open to investment”
8
. A study by CAPCO
showed that 38% of failed Hedge Funds were due to investment risk only while 54% were
due to operational issues. (See graphs below)
9
8
Edhec Risk and Asset Management Research, Jean-René Giraud, « Mitigating Hedge Funds’ Operational
Risks », June 2005
9
CAPCO, «Understanding and Mitigating Operational Risk in Hedge Fund Investments», March 2003
European Commission APPENDIX G
DG Internal Market
 128
Oxera (2006)
10
summarized the results of a survey conducted in 2003 where European
Asset Managers were asked to identify the most important operational risks the industry
faces and to evaluate their frequency and their potential impact. The study also highlighted
the direct link between the trend towards increased product sophistication and diversity and
the increase in the frequency or potential impact of these risks.
10
Oxera, «Current trends in asset management», 2006
European Commission APPENDIX H
DG Internal Market
 129
APPENDIX H: Risk Management Survey
1 Objectives
We conducted a risk management survey through an on-line questionnaire of 82 questions
covering all aspects of risk management. Respondents represent 55 asset managers,
including 22 managers of sophisticated UCITS and 15 managers of Funds of Hedge Funds.
Answers are presented for all respondents, but when the results are different for asset
managers of non-harmonised funds, this is specified and explained. This survey is
particularly important for understanding the impact of regulation and innovation in the
fund industry on risk management processes.
Where appropriate, we also included the results of face-to-face interviews conducted with
30 asset managers and professional associations. The summary of such interviews is
inserted as comment under the respective graphs, when it provide with some additional
details.
2 Results of the survey
2.1 Profile of the respondents
The survey respondents are predominantly Risk Managers, thus providing a valuable
source of information concerning risk issues. Along with CEOs, they represent more than
half of the total respondents. In fact, CEOs also are well represented, demonstrating an
active involvement within the investment process. The other category is mainly
represented by accountants and internal auditors.
Graph 1: Respondents by role, country of fund domiciliation and geographical distribution
14.3%
7.1%
2.4%
2.4%
7.1%
2.4%
40.5%
23.8%
0% 20% 40% 60% 80% 100%
CEO
Chief Operation Officer
Head of Investments
Head of Back Office
Business/Product Development
Manager
Portfolio Manager
Risk Manager
Other
Q1 - What is your role within the company?
European Commission APPENDIX H
DG Internal Market
 130
As regards geographical domicile and distribution, the majority of respondents have funds
domiciled in Luxembourg (74%) followed by France (42%). This does not reflect the
weight of these countries in the sample because the sample is based on fund domiciles. But
most funds selected for large European countries are managed by asset managers who also
have funds domiciled in Luxembourg. Among these funds, the greatest percentage is
distributed in the euro-zone or in the country’s own domicile. Surprisingly, the number of
funds distributed outside of Europe (e.g. USA, South America and Asia) exceeds the
number of funds distributed in European countries.
Face-to-face interviews confirmed this point. UCITS III funds are distributed in Asia and
South America where they are synonymous with investor protection and quality of
supervision.
Most of the respondents have more than 30 different funds offered to investors. Assets
under management ranged from €100 million to over €10 billion. As a matter of fact, 67%
of our respondents have assets under management of over €10 billion.
Asset Managers specialised in Non-UCITS, mainly Hedge Funds and Funds of Hedge
Funds, often have less than €10 billion under management.
Graph 2: Respondents by assets under management
42%
33%
33%
12%
74%
2%
2%
12%
28%
42%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
France
Germany
Ireland
Italy
Luxembourg
Poland
Slovakia
Spain
United Kingdom
Other
Q2 - In which of the following countries are your funds domiciled?
53%
84%
47%
79%
0% 20% 40% 60% 80% 100%
Extra Europe
Eurozone
Europe extra Eurozone
Country domicile
Q3 - In which countriesare your productsdistributed in?
14%
17%
12%
57%
0% 20% 40% 60% 80% 100%
1 - 5
5 - 15
15 - 30
> 30
Q4 - How many legal entities/fundsdoes your company
manage?
5% 5%
23%
67%
0%
10%
20%
30%
40%
50%
60%
70%
80%
100 - 500 mln
Eur
500 mln - 2.5
bln Eur
2.5 - 10 bln Eur > 10 bln Eur
Q5 - What is the total amount of assets under management of your
company?
European Commission APPENDIX H
DG Internal Market
 131
2.2 Investment products
In line with expectations, the respondents primarily distribute UCITS III funds, followed
by funds of hedge funds that seem to be preferred to hedge funds. Only 19% of
respondents solely distribute UCITS III products, while the others range from one type of
non-harmonised product, hedge funds or Fund of Hedge Funds in particular, to the full
range of non-UCITS funds.
Graph 3: Type of investment product and investor covered
On the other hand, all respondents have institutional clients. Other answers correspond to
funds that distribute their funds to retail and High Net Worth Individuals through
distributors (i.e. no direct access). In one case, the client is represented by sovereign
institutions.
It must be noted that UCITS III funds are extensively sold to institutional investors; some
UCITS III funds are even created only for the purpose of distributing them to pension
funds and insurance companies in Europe but also in Asia and South America.
Q6b: How many different type of funds are distributed
Ucits III and other
non harmonised
products
81%
Ucits III only
19%
97%
43%
59%
30%
27%
51%
0%
20%
40%
60%
80%
100%
120%
UCITS III Hedge Funds Fund of
Hedge Funds
Real Estate Private Equity Other non-
harmonised
f unds
Q6 - What type of Investment products are distributed by your company?
100%
73%
84%
57%
5%
0%
20%
40%
60%
80%
100%
Institutional Qualif ied Retail HNWI Other
Q8 - What type of inve stors do you cover?
European Commission APPENDIX H
DG Internal Market
 132
Graph 4: Preferred asset classes
The preferred asset classes to invest in are Equity and Fixed Income, where the former is
definitely the favourite asset class for the UCITS-only managers. Surprisingly, balanced
products are more heavily used than money market instruments, reflecting the increased
appetite of investors towards more rewarding investments. The same remark applies to
structured products.
Graph 5: Third party managed funds
A large share of respondents has funds managed by external fund managers. Among those
who distribute other funds than UCITS III only, the majority (86%) primarily outsources
UCITS funds, followed by Funds of Hedge Funds and Other non-harmonised funds.
97%
97%
76%
78%
76%
27%
0% 20% 40% 60% 80% 100%
Equity
Fixed Income
Money Market
Balanced
Structured Products
Other
Q7 - What type of asset classes are you investing in?
86%
14%
23%
5%
9%
27%
0% 20% 40% 60% 80% 100%
UCITSIII
HedgeFunds
Fund of HedgeFunds
Private Equity
Structured Products
Other non-harmonised funds
Q9b- If any, what type of investment products are managedbythirdparties?
3%
6%
58%
33%
0% 10% 20% 30% 40% 50% 60% 70%
Only
Mostly
Partially
No
Q9 - Do you have funds managed by third parties?
European Commission APPENDIX H
DG Internal Market
 133
Graph 6: Derivatives exposure in UCITS and non- harmonised funds
The use of derivatives still appears moderate (63%) within UCITS funds and almost 10%
of respondents show no exposure at all. The results are even lower for UCITS III-only fund
managers where the perceived use of derivatives is very moderate and the percentage of no
exposure is 33%. Surprisingly, even for non-harmonised investment funds, the use of
derivatives is quite limited. In any event, the portfolio analysis shows that the trend is to
have more derivatives in the portfolio, but even if there is an increase, the use of
derivatives remains moderate.
Graph 7: Derivatives asset classes and main reasons of exposure
The derivatives exposure is generally motivated by hedging purposes (92%) and almost
half of the respondents use derivatives for trading or arbitrage purposes. Other reasons
include, for instance, liquidity management or alpha porting.
The vast majority of funds utilises Equity and foreign exchange derivatives. Surprisingly,
the use of credit derivatives is extensive (67%). Other derivatives are mainly represented
by total return swaps or index swaps. In one case, we observed the use of swaptions and
volatility-based derivatives. UCITS-only managers also prefer Equity derivatives.
Face-to-face interviews confirmed the very low level of leverage used by European Asset
Managers. All interviewees confirmed that leverage for retail investors is not appropriate.
The portfolio composition analysis also confirmed the low level of leverage. It does not
appear to be a major risk in Europe for retail investors.
92%
56%
47%
11%
6%
0% 20% 40% 60% 80% 100%
Hedging
Trading
Arbitrage
Other
No exposure
Q14 - What is the objective of your derivatives exposure?
92%
83%
83%
28%
67%
25%
6%
0% 20% 40% 60% 80% 100%
Equity Derivatives
Interest Rate Derivatives
Foreign Exchange Derivatives
Commodity Derivatives
Credit Derivatives
Other derivatives
No derivatives
Q12 - What categories of derivatives are included in your portfolios?
30%
53%
7%
10%
0% 20% 40% 60% 80% 100%
Very Extensively or
Extensively
Moderately or Very
Moderately
No Exposure
No answer
Q11 - To what degree are you using derivatives in your non-
harmonised funds?
28%
63%
9%
0% 20% 40% 60% 80% 100%
Very Extensively or
Extensively
Moderately or Very
Moderately
No Exposure
Q10 - To what degree are you using derivatives in your UCITS
funds?
European Commission APPENDIX H
DG Internal Market
 134
Graph 8: Categories of derivatives
In line with expectations, the use of plain vanilla options and interest rate swaps is quite
extensive (over 80% of respondents). More complex derivatives are also used by numerous
respondents, especially structured products and credit linked notes (both at 69%). In one
case, variance swaps are also used.
2.3 Organization and Procedures
The number of people involved in the risk management unit can vary considerably, from
one to fifty persons, but, on average, the risk management unit is composed of 4 to
5 people.
Graph 9: Risk Management department
86%
83%
81%
69%
69%
47%
31%
6%
3%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Opt ion plain vanilla/ digit al options
Swaps (IRS, Cross-currency swaps, et c...)
FRA/ f orwards/ f ut ures
St ruct ured product s (basket linked, equit y
linked, caps/ f loors embedded, et c...)
Credit linked Not es/CDO/ CBO/ CDS
ABS/ CMBS/ Covered Bonds
More sophist icat ed opt ions (hybrid
opt ions, et c...)
No derivat ives
Ot her
Q13 - What type of derivatives/complex products are included in your portfolios?
45%
48%
6%
0% 20% 40% 60% 80% 100%
Yes
No
No answ er
Q16 - Has your Risk Management unit changed significantly after conversion to
UCITS III regulation?
European Commission APPENDIX H
DG Internal Market
 135
For almost half of the respondents (48%) the situation has not changed significantly after
the conversion to UCITS III, especially for those who have the largest risk management
team.
This statement also reflects the prudence of asset managers for which Risk Management
has always been important.
Graph 10: Risk Management Committee
Most of our respondents (68%) have a formal risk management committee. Both the Head
of Investments (32%) and CEO (39%) are well represented in the risk management
committee. Other front office roles, such as investment managers or other board members
are included in this committee.
The absence of Risk Management Committee does not mean that there is no Risk
Management function. All Asset Managers promoting sophisticated UCITS have a Risk
Management Committee. Asset Managers of non-harmonised funds rarely have a risk
management committee, but they do have a person in charge of risk management.
Obviously the size of the management company impacts the size and the scope of the risk
management unit.
Graph 11: Role and responsibilities of the risk Management Committee and role of the risk
management team
68%
32%
0%
20%
40%
60%
80%
100%
Yes No
Q18 - Do you have a Risk Management Committee?
32%
39%
18%
57%
11%
43%
25%
32%
0% 20% 40% 60% 80% 100%
Head of Investments
CEO
CFO
Head of Risk
Management
Res earch
Head of Compliance
Head of Legal
Other
Q18 - Who are the members?
81% 81%
50%
55%
64%
0%
20%
40%
60%
80%
100%
Approve new
products or
strategies
Approve risk
measurement
methodology
regarding new
products
Adjust risk
measurement
methodology in
case of market
Adjust exposure if
the level of risk
has breached an
alert point
Adjust exposure if
the level of risk
has breached a
limit
Q19 - What are the responsabilities of this Risk Management commitee?
96%
67%
67%
70%
65%
0% 20% 40% 60% 80% 100%
Product design
Strategy allocation
Tactical allocation
Day-to-Day risk position
taking
Client reporting
Q20 - Is risk management involved in the following steps
of the process cycle
European Commission APPENDIX H
DG Internal Market
 136
The Risk Management Committee is actively involved in the approval process of new
product strategies (81%) as well as in every decision related to risk measurement
methodology. From a technical side, the Risk Management team participates directly,
through a formal or informal validation process, in the product design (96%) and day to
day risk positioning.
Graph 12: Eligible products and authorisation process
As we can see, internally determined eligibility is an important criterion (75%) for the
inclusion of new products within newly-launched funds. Once the investments are
authorized, the funds are well monitored (77%) to ensure that the portfolio complies with
this eligibility requirement.
Asset Managers who have not defined any eligible investment products manage
non-sophisticated funds with simple investments. The same applies to Non-UCITS Asset
Managers for which eligibility is not a huge issue, as they work on the basis of more
flexible rules.
Graph 13: New Products Committee
75%
17%
8%
0% 20% 40% 60% 80% 100%
Yes
No
No answer
Q21 - Have you defined eligible investment products(e.g. derivatives)
for each fund ?
77%
20%
3%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
Yes
No
No answer
Q22: Is there a formal procedure to ensure that only authorized
investments/products are used ?
69%
28%
3%
0% 10% 20% 30% 40% 50% 60% 70% 80%
Yes
No
No answ er
Q24 - Do you have a committee for new products ?
74%
32%
26%
79%
26%
63%
53%
58%
26%
32%
0% 20% 40% 60% 80% 100%
Head of Inv es tments
CEO
CFO
Head of Ris k Management
Res ear c h
Head of Complianc e
Head of Legal
Head of Middle Of f ic e ( M/O)
Head of IT
Head of Bac k Of f ic e
Q24b - w ho a re the me mbe rs of the ne w products
commi tte e ?
European Commission APPENDIX H
DG Internal Market
 137
A new products committee has been established by the largest share of respondents (69%).
Risk managers are well involved in the approval process (79%) in cooperation with front
office personnel (Head of investments - 74%) and middle- and back-office roles.
Graph 14: Responsibilities of the new products committee
The main responsibilities of this committee consist in giving final approval to new
products (100% of respondents), while strategies, volume and limits are only approached at
this stage.
The following graphs explain the reasons that can lead to a new product being rejected
(41%).
Graph 15: Rejection of new products?
37%
100%
53%
53%
0% 20% 40% 60% 80% 100% 120%
Recommend new products
Provide f inal approval of
new products
Provide f inal approval of
new strategies
Provide recommendations
on volume and limits f or
new products and
strategies
Q25: What are the main responsibilities of this committee?
41%
48%
10%
0% 10% 20% 30% 40% 50% 60%
Yes
No
No answer
Q28 - Did you recently reject new products?
6%
19%
56%
19%
0% 20% 40% 60% 80% 100%
Impossibility of calculating
the risk
Impossibility of evaluating
products
Other (please indicate)
No answer
If yes, your reason was based on:
European Commission APPENDIX H
DG Internal Market
 138
The main reason for rejecting new products refers to the impossibility of evaluating the
new product (19%). Other reasons include:
 Difficult to monitor third-party investment managers;
 Impossible to reflect important product features in accounting system;
 Potentially difficult to market the products;
 Eligibility issues;
 Products inappropriately specified for markets or with insufficiently robust investment
processes;
 IT reasons.
The Committees which oversee new products may reject proposals for a number of
reasons, not all of which are linked to the use of derivatives. Such rejections may
encompass a number of issues, such as the lack of interest from distribution partners for
instance.
It is very interesting to see that asset managers do not hesitate to reject new products when
they feel that inappropriate valuation procedures or risk management procedures are
implemented. This can be explained by the fact that risk measurements provided by the
risk management unit are well used by front office (76%) in order to cope with the
different strategies conducted within the funds.
Face-to-face interviews confirmed that:
 The valuation risk, if we consider an increasing use of derivatives, appears to be
monitored closely and appropriately.
 The subprime crisis also shows that the valuation issue is not specific to complex
derivatives but should be assessed more effectively for more standard investments such
as the ones we find in money market funds.
Graph 16: Risk measurement
76%
10%
14%
0% 10% 20% 30% 40% 50% 60% 70% 80%
Yes
No
No answer
Q27 - Do Fund Managament and Front Office consider the results of risk
measurement helpful?
69%
14%
17%
0% 10% 20% 30% 40% 50% 60% 70% 80%
Yes
No
No answer
Q26 - Do you think, that your RM measurements appropriately reflect all
strategies performed (if applicable: arbitrage, long-short, ...)?
European Commission APPENDIX H
DG Internal Market
 139
3. Risk Features
As we can see on the graph below, all the potential risks to which a fund is subjected are
well covered by the risk management framework. In particular, between 60 and 80% of
respondents believe that Market Risk, Credit Risk and Operational risk are covered
appropriately.
Graph 17: Risk Framework coverage
A limited number of Asset Managers are working with derivatives on commodities. This
explains the low level of positive answers.
Q29 - Indicate and rate the risks covered by the risk management framework in
your company
43%
27%
30%
27%
37%
40%
27%
13% 13%
47%
43%
43%
37%
30%
33%
40%
37%
40%
7%
20% 13%
13%
27%
20%
30%
27%
27%
10%
3%
3%
3% 3%
0%
7%
10%
3%
10%
17%
3% 3% 3%
17%
7%
0%
20%
40%
60%
80%
100%
Market risk Liquidit y risk Count erpart y
risk
Credit Default
risk
(Operat ional
risk) Valuat ion
risk
(Operat ional
risk)
Compliance
risk
(Operat ional
risk) Ot her
Operational
risks
risk from
service
providers
risk f rom
int ermediaries
Very Good Good Medium Poor No Exposure
European Commission APPENDIX H
DG Internal Market
 140
28%
72%
0%
10%
20%
30%
40%
50%
60%
70%
80%
Yes No answer
Q30 - Do you feel comfortable with how the risk calculation of
commodity positions is calculated?
Graph 18: Differences between UCITS, non-UCITS and sophisticated UCITS funds
The respondents are equally divided about the risks arising from UCITS and Non-UCITS
funds. Almost half (47%) of the respondents believe that there is a difference for investors
between risks arising from UCITS and those arising from Non-Harmonised funds.
The interviews show that this difference is not linked to market risk but more to liquidity
risk. Often Non-UCITS funds are less liquid, as the redemption process is longer and
sometimes complex. Therefore, even if an investor understands perfectly market risk and
wishes to exit from non-harmonised funds, he/she may incur a loss as redemption notice
goes up to 3 months.
The second answer shows instead that the respondents believe that the risk profile of
sophisticated UCITS is specific, although some convergence is observed between
sophisticated UCITS and Non-UCITS funds (in particular Hedge Funds) in terms of
investment strategy. This convergence is not valid from a liquidity point of view as
sophisticated UCITS provide with daily liquidity which is not the case of non-harmonised
funds.
The subprime crisis leads us to qualify these statements. It must be noted here that most of
the UCITS which encountered liquidity issues because of the subprime crisis were not
sophisticated.
73%
13%
13%
0% 10% 20% 30% 40% 50% 60% 70% 80%
Yes
No
No answ er
Q32 - Do you consider that the risk profile of sophisticated UCITS funds is
different from that of non UCITS funds ?
47%
40%
13%
0% 20% 40% 60% 80% 100%
Yes
No
No answ er
Q31 - Do you consider the risks of UCITS funds as different from those of non
UCITS funds ?
European Commission APPENDIX H
DG Internal Market
 141
4. IT Tools and Data Management
As regards existing Risk Management systems, a majority of respondents (59%) are
confident about the ability of their systems to handle simple derivatives. More complex
products, on the other hand, are still far from being handled completely. Nonetheless, most
respondents (57%) feel certain that their systems will be capable of coping with new
products and strategies.
Graph 19: Ability of RM systems to cope with derivatives
Q34 - Do you consider your Risk Management systemas capable of
handling the following products and strategies?
59%
7% 7%
15%
26%
48%
33%
37%
7%
30%
26%
15%
7%
7%
11%
4%
26%
22%
4%
7%
0%
20%
40%
60%
80%
100%
Derivat ives - f irst generat ion
(swaps, caps/ Floors, simple equit y
opt ions, CDS, simple commodit y
derivat ives)
Derivat ives - second generat ion
(Spread opt ions, Basket opt ions,
CDO, Barrier options)
Hybrid product s (price depending
on dif f erent risk t ypes
simoult anously)
ABS / MBS
Fully Enough Partially No No answer
European Commission APPENDIX H
DG Internal Market
 142
Graph 20: Ability of RM systems to handle new products and strategies
Here again, we must insist on the fact that in practice, complex instruments are only
invested in when risk management and valuation procedures are ready.
Graph 21: Risk tools used by funds
11%
57%
25%
7%
0% 10% 20% 30% 40% 50% 60%
Fully
Enough
Partially
No answer
Q35 - Do you consider your system as capable of handling future new products
and strategies?
Q36 - Please indicate, which risk calculations are performed byeach system
61%
65%
37%
43%
46%
56%
44%
19%
23%
25% 15%
33%
25%
36% 16%
19%
37%
35%
11%
12% 26%
29%
14%
20%
37%
19%
8%
8%
0%
4% 4%
8%
0%
26%
35%
4%
0%
20%
40%
60%
80%
100%
VaR St ress t est ing Perf ormance
rat io
(Sharpe,...)
Tracking error Valuat ion Sensit ivit y
analysis
Backt est ing Count erpart y
risk
Credit Def ault
risk
External System: third party software Internal System: in-house developed
MS Office: excel spreadsheet, acces database Not Applicable
European Commission APPENDIX H
DG Internal Market
 143
As graph 21 shows, most respondents rely on external providers to calculate traditional
measures for market risk (VaR, stress tests, and sensitivity analyses), although some have
developed internal tools. On the other hand, credit risk is not outsourced and it is
calculated via in-house systems or basic tools, such as Excel spreadsheets. As we can see
in graph 22, the use of external systems and the development of internal systems have
significantly reduced the number of manual operations involved in risk calculation (only
37% of respondents). However, manual work remains important in the reporting process
(48%) and the valuation process (52%).
In general, all risk management operations are supported by an audit process (59% of
respondents).
Graph 22: Manual work and audit trail of RM Operations
59%
15%
26%
0% 10% 20% 30% 40% 50% 60% 70%
Yes
No
No answer
Q37 - Isthere an audit trail of all operationsperformed in the RMsystem?
52%
37%
48%
0% 10% 20% 30% 40% 50% 60%
valuation process
riskcalculation
reporting process
Q38 - Isthere significant manual work for:
European Commission APPENDIX H
DG Internal Market
 144
5. Valuation Process
Graph 23: Integration of new asset classes
A majority of respondents appear to be satisfied or very satisfied (50% and 14%
respectively) with the ability of their risk management systems to include new categories
of assets. This further confirms the amount of confidence placed by respondents in their
risk management department. In practice, when the answers are “medium” or “poor”,
Asset Managers prefer not to use new asset classes. Interviews confirmed that Asset
Managers are very careful before using any complex instruments.
14%
50%
25%
7%
4%
0% 10% 20% 30% 40% 50% 60%
Very Good
Good
Medium
Poor
No answer
Q39 - How would you rate the process of integrating a new asset class
into the Risk Management system?
European Commission APPENDIX H
DG Internal Market
 145
Graph 24: Source of valuation and pricing of derivatives
Graph 24 shows that the respondents do not repose too much confidence in a unique
pricing source for derivatives and other sophisticated instruments. Consequently, the use of
counterparty prices and those calculated via internal models is generally preferred.
The valuation risk does not appear to be significant for complex instruments. Valuation
issues are often considered carefully before accepting new investments. As shown by the
subprime crisis, valuation risk is more significant with investments which are deemed to be
standard.
Face-to-face interviews confirmed that there is a clever trend where investment banks are
creating increasingly complex products. They sell them extensively to asset managers for
structuring sophisticated UCITS but not always with a clear commitment to provide
liquidity and transparency in the valuation process. New products committees, as
mentioned in question 25, are particularly important in that context.
Q40 - To what extent do you rely on counterparty price ?
11% 14%
11% 11%
14%
11%
29%
14%
18%
21%
43%
36%
25%
29%
21%
21%
50%
39% 39%
14%
0%
20%
40%
60%
80%
100%
Derivatives - first
generation
(Swaps,
Caps/Floors,
simple equity
options, CDS)
Derivatives -
second generation
(Spread options,
Basket options,
CDO, Barrier
options)
Commodity
Derivatives
Hybrid
Instruments
ABS / MBS
Fully Mostly Partially Not used
Q41 - To what extent do you use internal models for valuation / pricing?
25%
18%
11% 11%
7%
18%
14%
11%
14%
14%
46%
43%
29%
32%
32%
25%
50%
43%
46%
11%
0%
20%
40%
60%
80%
100%
Derivatives - first
generation
(Swaps,
Caps/Floors,
simple equity
options, CDS)
Derivatives -
second generation
(Spread options,
Basket options,
CDO, Barrier
options)
Commodity
Derivatives
Hybrid
Instruments
ABS / MBS
Fully Mostly Partially Not used
European Commission APPENDIX H
DG Internal Market
 146
Graph 25: Validation of prices
* Such internal models pricing are developed by a dedicated risk management unit
independent from the unit managing transactions.
Furthermore, financial innovation allows more and more complex derivatives to be
created, especially OTC derivatives. Both answers below show that the respondents
recognize the difficulty of pricing new products.
Graph 26: Pricing issues for new investment products and non-UCITS.
Answers are similar for UCITS and non-UCITS. The valuation issue is recognized as a key
component of risk management processes as UCITS can invest in more and more OTC
derivatives, the level of difficulty of valuation converges with what was observable for
non-UCITS.
54%
32%
7%
4%
4%
0% 20% 40% 60% 80% 100%
Difficult
Medium
Easy
Very Easy
No answer
Q42: Would you say that being able to price the new investment productsis:
4%
41%
41%
4%
11%
0% 20% 40% 60% 80% 100%
Very Difficult
Difficult
Medium
Easy
No answer
Q43: Would you say that being able to price products used in non-UCITSis
75%
79%
68%
57%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
Counterparty price
Exchange traded price
Internal model price
Third parties price (Non-
Counterparty)
Q44 - Which valuation source do you use to control your prices
(especially for derivatives / complex instruments)?
*
European Commission APPENDIX H
DG Internal Market
 147
Graph 27: Internal Model validation
Again, the second answer of graph 27 shows that a majority of respondents repose great
confidence in the ability of the risk management unit to cope with pricing issues. In
addition, this confidence is often reinforced by a formal validation of the internal models
by an independent counterparty.
The importance of the pricing ability is related to the fact that for a majority of respondents
(59%) this represents a prerequisite for the trading activity of an instrument (see graph 28).
Graph 28: Trading requirements
Those percentages refer to all respondents.
If we consider asset managers of sophisticated UCITS, the percentage of positive answers
reaches 88%. It appears that for non-harmonised funds the ability to independently price an
instrument before trading it is not always a pre-requirement. The issuance of guidance by
international organisation like AIMA (Alternative Investments Management Association)
on pricing should help to improve this situation.
22%
22%
19%
11%
26%
0% 20% 40% 60% 80% 100%
All
Mostly
Partially
No
No answ er
Q45 - Are your internal models formally validated by an independent
unit ?
41%
37%
19%
4%
0% 20% 40% 60% 80% 100%
Fully
Mostly
Partially
No answer
Q46 - Do you consider that the Risk Management unit has an in-depth
understanding of the pricing models you use?
59%
41%
0%
20%
40%
60%
80%
100%
Yes No
Q47 - Is the ability to independently price an instrument a prerequirement
for trading it ?
European Commission APPENDIX H
DG Internal Market
 148
Graph 29: Flexibility and automation of the valuation process
4%
52%
41%
4%
0% 20% 40% 60% 80% 100%
Fully
Mostly
Partially
No answer
Q48 - How flexible is the valuation process to include a new complex
derivative ?
4%
43%
32%
14%
7%
0% 20% 40% 60% 80% 100%
Fully
Mostly
Partially
No
No answer
Q49 - Would you describe your pricing process as automated
(especially for derivatives/complex products)?
European Commission APPENDIX H
DG Internal Market
 149
6. Market Risk
Graph 30: Value at Risk
VaR is widely used, not only for sophisticated funds. It is an essential tool for Risk
Management units. Moreover, all the asset managers in our sample, managing only non-
UCITS, including Hedge Funds and Funds of Hedge Funds, measure the Value at Risk.
Graph 31: Value at Risk attributes
It is frequent for asset managers to use several methods to calculate VaR. This is explained
by the fact that Asset Managers recognize the impact of a method on the results.
68%
52%
52%
0% 20% 40% 60% 80% 100%
Historical
Parametric
Monte-Carlo
Q51 - Which method do you use to calculate VaR?
38%
8%
46%
8%
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
1 year 2 years > 2 years Other
Q52 - How old is your historical data series?
81%
19%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
Yes
No
Q50 - Do you assess VaR?
50%
50%
0% 20% 40% 60% 80% 100%
All UCITS f unds
only UCITS
sophisticated
Q50b - Do you assess VaRon UCITS funds?
European Commission APPENDIX H
DG Internal Market
 150
Graph 32: Value at Risk coverage
Those answers highlight the most important risk in the European Funds industry, i.e.
liquidity risk. Even if VaRs are properly calculated, the impact of liquidity is not always
assessed.
Graph 33: Conditional Value at Risk
Conditional VaR is used by asset managers applying some arbitrage strategies with an
extensive use of derivatives. They need strong risk management models for which
Conditional VaR is helpful.
54%
29%
18%
0% 20% 40% 60% 80% 100%
Yes
No
No answer
Q55 - Do you think, your VaR model sufficiently covers all your product
and portfolio risks?
4%
7%
21%
36%
32%
0% 20% 40% 60% 80% 100%
Fully
Mostly
Partially
No
No answer
Q54 - Does your VaR calculation take into account the liquidity of the
intruments/products?
31%
69%
0%
20%
40%
60%
80%
100%
Yes No
Q56 - Do you use conditional VaR?
European Commission APPENDIX H
DG Internal Market
 151
The use of back-testing in the context of risk management seems to be widespread, since
only 11% of respondents never perform back-testing. This is consistent with question 50
showing that 19% of respondents do not calculate the Value at Risk. In practice, back
testing is imposed by regulators who transposed the use of VaR for funds into their
regulation.
Graph 34: Back-testing
Graph 35: Back-testing reporting
25%
75%
0% 10% 20% 30% 40% 50% 60% 70% 80%
Yes
No
Q58 - Do you perform back testing on several confidence intervals?
41%
33%
15%
11%
0% 20% 40% 60% 80% 100%
All UCITS funds
Only UCITS sophisticated
funds
Non UCITS funds
Never
Q57 - Do you perform a backtesting for:
37%
63%
0%
20%
40%
60%
80%
100%
Yes No
Q60 - Do you include backtesting results in your reporting ?
European Commission APPENDIX H
DG Internal Market
 152
Graph 36: Stress-testing
Graph 37: Traditional ratios and other risk measures
The Sharpe ratio is used across the industry even if it is frequently criticised for not being
appropriate. It is also used extensively for non-UCITS. For Funds of Hedge Funds, the
maximum drawdown is always used, but with other, more traditional ratios.
Graph 38: VaR model Validation
50%
62%
19%
31%
0% 20% 40% 60% 80% 100%
Risk f actor correlation
Risk factor standard
deviation
Liquidity
No stress test
Q62 - If you stress test VaR parameters, what do you stress?
48%
11%
30%
11%
0% 20% 40% 60% 80% 100%
All UCITS f unds
Non UCITS f unds
only UCITS sophisticated
funds
Never
Q61 - Do you performStress testing for:
82%
14%
7%
46%
61%
29%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
Sharpe ratios
Sortino Ratio
Omega Ratio
Maximumdraw down
Sensivity Analysis
Other
Q63 - Do you use the following ratio/risk measures for your funds?
56%
19%
26%
0% 20% 40% 60% 80% 100%
Yes
No
No answer
Q64 - Have your VaR models been validated by an independent unit?
European Commission APPENDIX H
DG Internal Market
 153
7. Other Risks
Graph 39: Credit risk
Credit risk is considered a significant risk and appropriate methods are implemented on the
basis of regulations. In practice, the limits applied by asset managers are sometimes more
restrictive than those set forth by regulators.
Graph 40: Liquidity Risk
The liquidity analysis is developed and applied but not to a large extent. Question 54
already showed that VaR calculation does not always include the impact of liquidity. This
is even more the case for non-harmonised funds.
81%
19%
0%
20%
40%
60%
80%
100%
Yes No
Q65 - Do you have measurement methods and limits for Credit risk?
33%
37%
30%
0% 5% 10% 15% 20% 25% 30% 35% 40%
Yes
No
No answer
Q66 - If yes, are your limits more restrictive than those given by your
national regulation authority?
19%
15%
41%
26%
0% 20% 40% 60% 80% 100%
A ll
Mos tly
Partially
No
Q67 - Do you include a liquidit y analysis of authorize d pr oducts?
European Commission APPENDIX H
DG Internal Market
 154
Graph 41: Counterparty risk
Counterparty risk is logically measured by most asset managers, as well as collateral
agreements. This is imposed by UCITS III for OTC derivatives. Counterparty risk
measurement tends to be harmonised among asset managers even if it is not harmonised by
regulators as shown by the comparative Regulatory Analysis. Asset Managers tend to
adopt the most conservative approach when performing measurements.
Graph 42: Operational risk
The quantification of operational risk by the asset managers in our sample clearly shows
that key risks are identified properly and that an appropriate Risk Management framework
is in place. Moreover, the results of operational risk analyses are routinely reported to
senior management.
79%
21%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
Yes
No
Q68 - Do you measure Counterparty Risk for derivatives?
81%
19%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
Yes
No
Q69 - Do you use collateral agreements to reduce Counterparty
exposure?
75%
59%
79%
75%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
set up a global operational
risk framework
performa risk cartography
perf orma risk Self
Assessment
collect losses in a database
Q71 : What are the activities of your Operational risk framework?
81%
19%
0%
20%
40%
60%
80%
100%
Yes No
Q70 - Do you quantify your Operational Risks?
European Commission APPENDIX H
DG Internal Market
 155
Graph 43: Operational risk reporting
Graph 44: Operational risk assessment
Graph 45: Operational risk of UCITS and non-UCITS funds
88%
84%
84%
76%
70% 72% 74% 76% 78% 80% 82% 84% 86% 88% 90%
Quality of internal controls
Non-compliance w ith
f und's investment
policy/f und prospectus
Non-compliance w ith
external regulation
Valuation risk
Q73 - Does your operational risk assessment include?
7%
25%
32%
29%
7%
0% 20% 40% 60% 80% 100%
Fully
Mostly
Partially
No
No answer
Q74 - Do you consider that new investment powers (UCITS III) have
increased your operational risk?
7%
32%
43%
18%
0% 20% 40% 60% 80% 100%
Fully
Mostly
Partially
No
Q75 - Do you consider that Non-Ucits/Hedge Funds are subject to an
untypical high operational risk?
86%
14%
0%
20%
40%
60%
80%
100%
Y es No
Q72 - Do you re port the re sul ts of the ope ra ti ona l ri sk a na l ysi s on a re gul a r ba si s to
se ni or ma na ge me nt?
European Commission APPENDIX H
DG Internal Market
 156
UCITS III has a significant impact on operational risks, as a direct result of the new
investment powers granted under the directive. This point was clearly confirmed during
face-to-face interviews. Non-UCITS, in particular Hedge Funds, are still perceived as
being subject to atypically high operational risks.
This atypically high operational risk is linked to valuation and liquidity issues. It is
important to note that the answers to this questionnaire were received before the subprime
crisis broke out.
8. Reporting
Graph 46: Overall risk management reporting
100%
77%
73%
93%
89%
81%
64%
89%
81%
69%
85%
67%
85%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Market Risk
Credit risk
Count erpart y risk
Compliance wit h invest ment policy
Compliance with regulat ion
Int ernal control risks
Valuat ion risk
VaR
St ress-t ests
Result s of validat ion work/ Back t ests
limit breaches/ point s of alert
Qualit at ive comment s on risks
Execut ive summary wit h main risks and main events
Q76 - Does the risk management reporting cover:
European Commission APPENDIX H
DG Internal Market
 157
Graph 47: Reporting preparation
The quality of reporting is so important that validation by the risk management unit is
often required.
Graph 48: Reporting appropriateness
As the above answers show, asset managers are aware that, to a certain extent, reporting
packages can be improved to provide an accurate picture of risks.
75%
25%
0%
20%
40%
60%
80%
100%
Yes No
Q77 - Is all reporting prepared or validated by the RM unit ?
32%
50%
14%
4%
0% 20% 40% 60% 80% 100%
Fully
Mostly
Partially
No answer
Q78 - Do you consider that your reporting package provides a good
picture of risks you face?
European Commission APPENDIX H
DG Internal Market
 158
9. UCITS III and new products development
Graph 49: UCITS III funds
As confirmed by interviews, the transition to UCITS III has been a slow process. In
practice, most asset managers did use new investment powers such as funds of funds or
cash and money market funds, but they waited for classifications by regulators before
launching sophisticated UCITS.
Graph 50: UCITS modifying directives improvements
85%
89%
15%
0% 20% 40% 60% 80% 100%
UCITS f unds launched after
2002
UCITS f unds launched
before 2002, but converted
during the transitory period
UCITS f unds launched
before 2002, but converted
after the transitory period
Q79 - Do you have UCITS III compliant funds in your company?
78%
37%
70%
44%
4%
0% 20% 40% 60% 80% 100%
Investments Power (eligible
assets)
Fund Perf ormances
Risk Prof ile
Geographical Distribution (sales)
No improvement
Q80 - What areas have the UCITS III directives improved?
European Commission APPENDIX H
DG Internal Market
 159
The global impact of UCITS III is seen as positive. According to the asset managers in our
sample, UCITS III has impacted investment powers and risk profiles, even if performance
analyses showed that such impact is neutral. This is due to the fact that most sophisticated
funds were launched in 2007.
The perceived impact on risk profile is crucial. This perception is linked to the fact that
investment powers have become so important that the suitability of products to the
investment profile is now essential. Some UCITS III are created for institutional investors
only and are not suitable for retail investors.
Graph 51: Current and anticipated changes
Organisations had to adapt to UCITS III as well as processes linked to Risk Management.
This trend will continue in the next twelve months. Product innovation plays a key role in
the European Investment Fund industry and most players recognize that they do not take
advantage of all the flexibilities available under UCITS III. To a certain extent, Asset
Managers specialising in the Hedge Fund business would like to benefit from a European
Passport. This is more obvious for those specialising in Funds of Hedge Funds but the
liquidity of such products should be dealt with carefully.
52%
78%
70%
22%
35%
0% 20% 40% 60% 80% 100%
Organisational
Process
Products
IT
Risk Management Methodology
Q81 - Has the implementation of UCITS III in your country generated changes in your
company?
48%
65%
74%
61%
52%
0% 20% 40% 60% 80% 100%
Organisational
Process
Products
IT
Risk Management Methodology
Q82 - Do you foresee implementing any upcoming changes or improvements within
your company in the next 12 months?
European Commission APPENDIX I
DG Internal Market
 160
APPENDIX I: COMPARATIVE REGULATORY ANALYSIS – ADDITIONAL
INFORMATION.
European Commission APPENDIX J
DG Internal Market
 161
APPENDIX J: Alternative Investment Survey
European Commission APPENDIX J
DG Internal Market
 162
European Commission APPENDIX J
DG Internal Market
 163
European Commission APPENDIX J
DG Internal Market
 164
European Commission APPENDIX J
DG Internal Market
 165
European Commission APPENDIX J
DG Internal Market
 166
European Commission APPENDIX J
DG Internal Market
 167
European Commission APPENDIX J
DG Internal Market
 168
European Commission APPENDIX J
DG Internal Market
 169
European Commission APPENDIX J
DG Internal Market
 170
European Commission APPENDIX J
DG Internal Market
 171
European Commission APPENDIX J
DG Internal Market
 172
European Commission APPENDIX J
DG Internal Market
 173
European Commission APPENDIX J
DG Internal Market
 174
European Commission APPENDIX J
DG Internal Market
 175
European Commission APPENDIX J
DG Internal Market
 176
European Commission APPENDIX J
DG Internal Market
 177
European Commission APPENDIX J
DG Internal Market
 178
European Commission APPENDIX J
DG Internal Market
 179
European Commission APPENDIX J
DG Internal Market
 180
European Commission APPENDIX J
DG Internal Market
 181
European Commission APPENDIX J
DG Internal Market
 182
European Commission APPENDIX J
DG Internal Market
 183
European Commission APPENDIX J
DG Internal Market
 184
European Commission APPENDIX J
DG Internal Market
 185
European Commission APPENDIX J
DG Internal Market
 186
European Commission APPENDIX J
DG Internal Market
 187

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close