alfi response to esma’s discussion paper on ucits share · alfi response to esma’s discussion...

Luxembourg, 27 March 2015

ALFI response to ESMA’s Discussion Paper on UCITS share

classes

General Remarks

The Association of the Luxembourg Fund Industry (ALFI) is the representative body of the Luxembourg

investment fund community. Created in 1988, the Association today represents over 1400 Luxembourg

domiciled investment funds, asset management companies and a wide range of service providers such

as custodian banks, fund administrators, transfer agents, distributors, legal firms, consultants, tax

experts, auditors and accountants, specialist IT providers and communication companies. The

Luxembourg Fund industry is the largest fund domicile in Europe and a worldwide leader in cross-

border distribution of funds. Luxembourg-domiciled investment structures are distributed on a global

basis in more than 70 countries with a particular focus on Europe, Asia, Latin America and the Middle

East.

It must be underlined that the Luxembourg fund industry is in a specific situation with regard to the

creation of share classes since it has an exposure to a certain number of distribution countries and as

a result has to adapt to a wide variety of investors’ needs. ALFI therefore welcomes the opportunity to

respond to the European Securities and Markets Authority’s (“ESMA”) discussion paper on share

classes of UCITS.

Questions

1. What are the drivers for creating different share classes?

Share classes are mainly launched to respond to distribution needs. In particular, it helps expanding

the client base globally (i.e. share classes offered in additional currencies,…), meet new client demands

(i.e. requests for specific hedges, for example duration hedges), or meet new regulatory demands (i.e.

inducements free share classes). They also give asset managers the possibility to offer products that

are aligned to the investment profile of investors (appetite for dividends, a specific currency, their own

asset/liability management in terms of currencies, interest exposure, timing of liabilities,…).

ALFI Response to the ESMA Discussion Paper on share classes of UCITS

There are cost reasons for the development of share classes too: the creation of a sub-fund involves

set-up costs and, depending on the fee structure, may involve additional operational costs which

may penalise sub-funds with low levels of assets under management (e.g. transaction costs or when

minimum fees or fixed fees apply). The creation of a separate share class is thus in the interest of

the investors.

Creation of share class may also give access to hedging features that protect investors. Without the

possibility to hedge at share class level, investors might not have access to such hedging features,

as hedging outside of the fund is not affordable for smaller investors.

Related operational aspects are also to be noted: the investment manager may not be able to co-

manage assets belonging to different sub-funds.

Moreover the seeding money aspect is not to be neglected: a sub-fund requires higher assets under

management (min. EUR 15 million) than a share class. Launching a sub-fund thus requires that

asset managers tie up more capital in the form of seed money as well as having to make additional

sales efforts versus creating a share class.

Another advantage of share classes is time to market: a sub-fund must be approved by the

supervisory authority prior to its launch. A prospectus revision requires time (in particular, in terms

of drafting and review by the supervisory authority) and may entail delays whereas a share class can

be launched on an as-needed basis.

It is useful to note that in certain non-EU countries, in particular Asia, sub-funds may only be

launched subject to certain conditions of size and NAV track record, which may take a long time.

Share classes are therefore useful to launch products on the market within a reasonable timeframe.

It must also be underlined that the Luxembourg fund industry is in a specific situation with regard to

the creation of share classes since it distributes UCITS to investors in a large number of distribution

countries in and outside of the EU. Therefore it has to adapt to a wide variety of investors’ needs.

Up to now we have never faced any issue with our share class range.

Finally, the creation of share classes responds to the goals of the European Commission’s Green

Paper on the Capital Markets Union, by ensuring a wider choice among available products for

investors, lowering prices and facilitating cross-border retail participation in UCITS.

2. Why do certain UCITS decide to create share classes instead of setting up a new UCITS?

As mentioned above in our response to question 1, the creation of share classes allows for the

reduction of the overall costs of umbrella sub-funds and economies of scale when the expected

assets under management are not sufficient to manage a sub-fund in an economically efficient

manner (because of administration costs, agent’s maintenance of accounting records, compliance,


risk management, audit etc.) or in case of an urgent investment demand. Moreover this reduces the

go-to-market time of new products.

3. What are the costs of creating and operating a new share class compared to the cost of

creating and operating a separate UCITS?

Costs of creating and operating a new share class are obviously much lower. It is however difficult

to give a precise figure since such costs can vary depending on the size, the targeted markets and

distribution countries and the type of structures concerned. It must be noted that there is a certain

unavoidable amount of fixed costs incurred when creating a sub-fund, such as set up costs,

registration, accounting, or audit fees. Most of these costs are not required when setting up a share

class or are generally limited.

4. What are the different types of share class that currently exist?

In addition to the share classes mentioned in the ESMA Discussion Paper, we may cite:

Founder share class: a class reserved to the investor which has invested during the initial

subscription period and which is not available to other investors.

Seeding share class: a class with a lower management fee but with specific minimum initial

investment amount and/or minimum holding amount requirements as determined by the board of

directors and which remains open until the sub-fund has reached a critical size in terms of assets

under management or a decision of the board is taken to close the share class based on reasonable

grounds.

Finally there are also share classes that provide hedging, such as for instance duration hedging to

either remove interest rate risk or reduce it in a specific target bandwidth specified in the prospectus,

or provide Beta hedging in order to let the investors choose if they want to hedge any Beta exposure

to the market (Delta/Market hedged share classes).

5. How would you define a share class?

Share classes / unit classes in a UCITS are categories of shares which belong to the same UCITS

sub-fund and which all confer the same rights to shareholders/unitholders of the UCITS and allow

them to achieve some level of customisation which accommodates their specific needs and risk

appetite. In this respect, we agree with the ESMA Discussion Paper.


A share class is a product which should have the same investment policy than other share classes

of the same fund or sub-fund. However, the performance of each share class can differ from another

due to its set of characteristics such as fees and reduced exposures due to hedging.

We agree with the principle that one share class’s specific features should not cause material

adverse impacts for holders of different share classes. It must be underlined that Luxembourg UCITS

management companies have already put robust systems of control in place to mitigate such risk,

as already required by the Luxembourg regulator.

As to the final point under paragraph 6 of the Discussion Paper, we would note that differences

between share classes are already adequately disclosed in the fund documents, in particular in the

KIIDs where each share class has its own KIID.

6. Do you agree that share classes of the same UCITS should all share the same investment

strategy? If not, please justify your position.

We agree with the principle that share classes of the same UCITS sub-fund should all share the

same investment strategy to the extent that, in the absence of a harmonised EU definition of this

notion this should mean that there is a common pool of assets reflecting the policy of the relevant

UCITS sub-fund. Therefore specific justifiable differences should be allowed on top of this same

investment strategy to protect the investors from negative impacts such as currency exchange risk

and interest rate risk.

More specifically, in relation to sections 9 and 11 of the Discussion Paper, it must be noted that

none of the provisions of the UCITS directive excludes the eligibility of a share class hedging the

interest rate risk. We do not share the view that these instruments are an investment asset of the

fund but consider them rather as instruments used with a view to implementing acceptable hedging

techniques.

Indeed, an interest rate hedged share classes aims to systematically hedge out a specific risk like a

currency hedged share class does. It does not have an adverse impact on shareholders in other

share classes as it bears its own costs of hedging (separate P/L reflecting specific hedging costs).

Differences between the share classes within a fund are also clearly disclosed to investors.

It must be underlined that such share classes were never authorised without the assurance that the

interests of investors of other share classes were protected.

7. Could you explain how the operational segregation between share classes works in

practice?

Different methods are used.


The administration agent and investment manager apply an operational and accounting segregation

at the level of each share classes. As a result:

(i) the P&L (including the performance) is distinct for each share class; and

(ii) the costs linked to a share class are allocated to the relevant share class (e.g. brokerage

fees, performance fees).

(iii) The operational and accounting segregation is maintained at all times: the administration

agent ensures that the transactions made in the context of the policy applied to a share class are

identified and allocated to a bespoke technical account which is linked to the same share class.

The bespoke technical account is set up both at the level of the investment manager and

administration agent. The investment manager ensures that it has, at all times, sufficient liquidities

(cash and/or liquid instruments) in order to honour margin calls in relation to the hedging

techniques. The use of bespoke technical accounts permits an operational segregation of such

flows and a proper allocation of the costs to the relevant P&L.

Operational segregation obviously goes with the organisation of appropriate risk management

processes.

Finally, from an investor protection perspective an ISIN code per share class can be allocated.

8. Do you agree that the types of share class set out in paragraph 8 are compatible with

the principle of having the same investment strategy? In particular do you agree that

currency hedging that is described in paragraph 8 complies with that principle? If not,

please justify your position.

We agree. However currency hedging as described is much too limited. It should also include hedging of currency risk at underlying portfolio level. We also believe that the share classes offering hedging features such as duration hedging, and other types of hedging should be allowed as well.

9. Do you believe that other types of share class that comply with the principle of having

the same investment strategy exist (or could exist) and should be allowed? If yes, please

give examples.

As mentioned above, duration hedged share classes should be allowed since they have the same

investment policy and follow the same principles as the currency hedged share classes. Hedge

share classes offer the opportunity for investors to have access to hedging techniques which they

would not have via another type of investment structure, and that may not be accessible for them

outside of the fund structure.


The portfolios of both types of share classes have the same investments as the non-duration hedged or non-currency hedged share classes, respectively, except that, in addition to these investments, they both use derivative instruments in order to hedge either the duration risk or currency risk. Financial derivative instruments (such as futures or swaps) are used for risk reduction purposes only (as opposed to investment purposes). Both classes achieve the same exposure to the market as the non-duration hedged share classes and the non-currency hedged share classes. They thus comply with the requirement of having the same investment strategy. Finally, they allow for a reduction of costs. Banning such share classes would result in having to create sub-funds which would have too small assets and this would be much more expensive for the investors.

As such small funds would not be viable, and thus not be launched, the expected outcome would

likely be that such hedging features would not be available anymore to investors.

In consideration of the above, we do not see a reason to allow currency hedged share classes

but to exclude duration hedged share classes. ALFI would suggest that focus is instead put on

ensuring that appropriate risk warnings are inserted in prospectuses and information on

transactions at share class level is more detailed in annual reports.

10. Do you agree that the types of share class set out in paragraph 10 above do not comply

with the principle of having the same investment strategy? If not, please justify your

position.

For the reasons explained under paragraph 9 above, share classes that offer differing degrees of

protection against interest rate risk (i.e. duration hedged share classes) comply with the principle

of having the same investment strategy. The goal of the proposal is to only allow standard

differentiation features between share classes and forbid different exposures that would lead to

different “results of the investment strategy”. Other products such as duration and Beta hedged

share classes have a real beneficial hedging effect for their investors to protect them against an

interest rate rise or negative performance of the overall market (i.e. to follow the principle of

hedging protection).

We therefore disagree with considering as not compliant per se share classes that offer differing

degrees of protection against some market risks such as interest rate or volatility risk, and share

classes that are exposed to the same pool of assets but with different levels of capital protection

and/or payoff. It must be noted that existing share classes based on that profile have been set up

to respond to investors’ requests and their investment profiles, and are always authorised under

strict conditions.

Moreover, ALFI is of the view that there should be no exhaustive list of compatible share classes

since the creation of share classes responds to requests from investors. It is essential that the

asset management industry be able to continue to create products in line with investors’


investment profiles and needs. A case by case assessment when proposed new share classes

are reviewed by authorities based on the ESMA principles should be sufficient.

11. Please provide information about which existing UCITS do not comply with the criteria

laid down in paragraph 6 as well as an indication of the assets under management and the

number of investors of these UCITS.

The vast majority of UCITS do already comply with the criteria laid down in paragraph 6. However

we note there are share classes duly authorised, whilst being fully compliant with paragraph 6,

that are not mentioned in paragraph 8, as this list is not exhaustive.

12. Do you see merit in ESMA clarifying how regulatory ratios such as the counterparty

risk limit should be calculated (e.g. at the level of the UCITS or share classes)?

In our view the existing ratios applied at the level of the UCITS sub-fund should remain and are

sufficient; information available in the prospectus is already comprehensive. Existing provisions

have indeed put appropriate quantitative rules and risk has been adequately addressed over the

years by the industry. Introducing new obligations in this regard would add to existing costs.

Furthermore we see no need to clarify how ratios should be implemented since there is no unique

solution in this context that would fit all situations. As mentioned above, when share classes are

authorised it is always on the condition that risks are identified in the prospectus, and appropriate

risk management processes are put in place.

13. Do potential and current investors get adequate information about the characteristics,

risks and return of different classes in the same UCITS? If not, what else should be

provided to them?

In our view, fund documents such as prospectuses and KIIDs provide sufficient information. Best practice in this context encourages transparency in the fund documents. ALFI would like to underline ESMA’s good work regarding its discussion paper on PRIIPs in this regard. The basis for providing adequate information about the classes of UCITS therefore already exists.

14. Do you agree that ESMA should develop a common position on this issue? If not,

please justify your position.

ALFI understands ESMA’s concerns to bring clarity on what a share class is and the ways in

which such share classes can differ between one another. However such an initiative should not


restrict the use of share classes which provide a cost-effective way of ensuring UCITS meet the

needs of investors and are aligned with investors’ investment profiles. Too detailed rules wouldn’t

make sense in our view, as the creation of share classes should typically be subject to a case by

case assessment by the supervisory authority concerned.

Banning a large amount of share classes would lead to further fragmentation in the market and

the risk of launching too many sub-funds, thereby adding new fees for investors. This would be

contrary to the spirit of the Capital Markets Union as outlined in the recent European

Commission’s Green Paper and in particular to the need to consolidate investment funds in

Europe.

In our view existing share classes should not be impacted by any decision taken by ESMA. Some

of these classes are indeed small and the fund promoter may not have any other choice than

simply withdraw them without launching any new classes if they are banned. This would not be

to the benefit of the investors in these classes. In addition, in certain markets this would create

additional issues for example, in the specific situation of share classes distributed in Taiwan. As

mentioned above in our response to question 1, such a problem would be even more substantial,

since the newly created sub-fund would not be eligible for registration for a significant length of

time due to the 3 year track record requirements.


Appendix: Identification of the stakeholder

• Association of the Luxembourg Fund Industry

12, rue Erasme L-1468 Luxembourg Contact person for this consultation: Evelyne Christiaens ([email protected])

• If you are registered with the Commission as an "interest representative" your identification number:

6182372280-83

• Are you a recognised European social partner organisation or a representative of a European (sectoral) social dialogue committee?

No.

• Field of activity of the respondent. Please specify your field of activity. Please indicate if you are directly affected by any of the measures and if so, which one and to what extent:

ALFI is the official representative body for the Luxembourg investment fund industry. Being a non-profit organisation, ALFI is not directly affected by any of the measures.

• If the respondent is an association of stakeholders, how many members do you represent and what is your membership structure?

ALFI represents over 1188 investment funds, 247 service provider and 19 associate members.

• Do you object to publication of personal data on the grounds that such publication would harm your legitimate interests?

No.

• Do you agree to having your response to the consultation published along with other responses?

Yes.

alfi response to esma’s discussion paper on ucits share · alfi response to esma’s discussion...

Documents