Mifidcesr1stmand2ndconsresponsefinal.pdf

International Swaps and Derivatives Association
International Securities Market Association
International Primary Market Association
Association of Norwegian Stockbroking Companies
Bankers and Securities Dealers Association of Iceland
Bond Market Association
Danish Securities Dealers Association
Finnish Association of Securities Dealers
Futures and Options Association
London Investment Banking Association
Swedish Securities Dealers Association
Joint Response to CESR’s November 2004 consultation on Draft Technical
Advice on Possible Implementing Measures of the Directive 2004/39/EC on
Markets in Financial Instruments (MIFID, 1st Set of Mandates, 2nd
consultation)
17 December 2004
CONTENTS

Introduction
1. General questions from the first consultation 3. Record keeping and the burden of proof 6. Conflicts of interest and the segregation of areas of business 8. Methods and arrangements for reporting financial transactions 9. Most relevant market in terms of liquidity 10. The minimum content and the common standard or format of the reports to facilitate its exchange between competent authorities Annex. MIFID provisions where transitional measures are most needed. Page 16 INTRODUCTION
This response is a compilation of the agreed views of the eleven associations listed at the beginning. For the purposes of its analysis of responses CESR should count this response as coming from eleven respondents and weight it accordingly. The associations represent a significant proportion of investment firms active in the European securities and derivatives markets, especially its wholesale markets. Our response follows the order of the CESR consultation paper. Except for sections 8 to 10 on transaction reporting, CESR has not consulted on revised articulated text. It would have been helpful to have an opportunity to comment on CESR’s revised draft text, both in these areas and where CESR has decided its proposed approach. Our comments in response to CESR’s earlier consultation remain highly relevant. In the absence of specific text it is not possible for us to provide definitive comment on CESR’s revised proposals. This makes it particularly important that CESR’s advice should take the more general, less detailed and prescriptive form that we and other commentators recommended, given the danger of technical problems if CESR were to publish very detailed advice, and the fact that Level 3 gives more scope for the modification and update of detailed regulatory provisions. General questions from the first consultation

(c) Transitional issues
We attach in the Annex a schedule of the provisions where transitional measures are
most needed, together with a short narrative describing how, typically, a large
investment bank might set about a systems project of the kind that will be required by
some of the MIFID provisions. The schedule does not cover extensive work that will
be required for many other small adjustments required under MIFID. The timetable-
critical work is in addition to this extensive, but less time-critical, ‘other’ work. The
narrative draws on firms’ experience of project managing major system changes. It
aims to illustrate the real time-line for systems development of this kind.
The schedule also includes notes on MIFID provisions that will be particularly costly
for firms to implement (in addition to the cost of the systems changes mentioned
already).
In addition to widespread concerns that firms will be unable to implement changes
within the current timetable, and that in consequence they will be exposed to liability
in their dealings with clients (as well as regulators), perhaps an even greater concern
is that firms’ inability to deliver by the given date might affect market and financial
stability. The particular concerns are that:
It is the biggest firms that are most exposed as, whilst they have the greatest systems resources, equally they operate the most complex systems, have the largest and most international client base, and therefore carry a high level of regulatory risk and a substantial burden of documentation. If some of these firms chose to withdraw certain services because of their liability to clients under new rules with which they were not yet able to comply, the ability of the market to conclude trades would be seriously impaired. Authoritative market share estimates are notoriously difficult to obtain. However, a 2003 survey of wholesale cash equity transactions by a large number of the largest houses suggested that that those firms accounted for approximately 60% of cash equities trading in the London market and some 30% respectively in Frankfurt and on Euronext. As these figures exclude retail operations, and some of the houses have significant retail presence, the percentages would increase if this were taken into account. This reduction in service, if it were to happen, would act to destabilise the market pricing mechanism. The impact would be felt by institutions and by consumers alike. Even if, on the other hand, these firms chose to operate services in full, but implemented MIFID late (an unlikely scenario, since firms would be unlikely to accept the regulatory liability they would face be continuing rather than ceasing business in areas where they were not yet in compliance), their collective exposure to civil liabilities to clients could introduce a serious contingent liability. This would attach solely to the providers of liquidity. The amount of risk capital that would be required to be set aside to deal with this potential liability could, in itself, serio usly impair the provision of liquidity to the market. Independence of compliance

It is not entirely clear what CESR means when it refers to a functional, rather than an
organisational perspective, where independence of compliance is achieved through
the "four eyes" approach. This may be welcome if it means that one of two directing
individuals in a firm can assume the overall control of the compliance function so
long as he acts independently in doing so (and even though he might be involved in
some business activities).
However, in general, we would prefer CESR to adopt a generalised version of the
proposal in sub-paragraph (ii) in the last bullet point of this section. All firms,
regardless of size, should have flexibility as regards the means to achieve the
objective which the requirement for an independent compliance function is intended
to attain, i.e. that the firms have in place adequate systems and controls designed to
ensure effective compliance, having regard to the nature of the firm's activities. A
rigid requirement for either outsourcing or independence would be inappropriate and
impose excessive cost, for example, for firms providing only advice, or firms only
operating an MTF. Furthermore, in a small firm, the senior management’s closeness
to the business may enable more effective compliance than could be provided by an
independent or outsourced function. In all cases, compliance should be treated as a
culture which infuses the whole firm, and not as a separate function to be hived off
into a separate and remote unit.

3.
Record keeping and the burden of proof
We welcome CESR’s intention not to reverse the burden of proof. However, our view remains that CESR’s proposals as set out in its first consultation paper on the first set of mandates do not achieve this intended result. CESR's original proposals went beyond mere record keeping obligations e.g. paragraph 7(b) Box 7 would allow the regulator to impose fines if in it deems the firm's records insufficient to demonstrate compliance, whether or not there is in fact a failure to comply. In addition, it seems from CESR's paper that it also wishes to see generalised requirements on firms to keep detailed records which demonstrate compliance with each and every requirement of each and every rule which applies to a firm. Generalised record keeping obligations of this kind would effectively reverse the burden of proof. They would enable the regulator to avoid having to show that the substantive rule has been broken before imposing sanctions. If the firm could not produce sufficient records to demonstrate compliance with the substantive rule, the regulator would simply impose sanctions for breach of the record keeping requirement, without the need for the regulator to show that the firm had in fact contravened the substantive rule. In fact, generalised record keeping obligations are more stringent than a simple reversal of the burden of proof. If the firm were to have the burden of proof of showing that it had in fact complied with a substantive rule, it could discharge that burden by calling witnesses and proving compliance by oral testimony (depending on
the legal framework in the member state concerned). In contrast, if there were
generalised record keeping obligations, the only way the firm could escape sanction
would be to produce contemporaneous documentary records that completely
demonstrated that it did what was required.
As we indicated in our earlier submission, we consider that this effective reversal of
the burden of proof would go against the decision made at Level 1 and may
contravene constitutional requirements of a number of member states.
The breadth and range of some of the rules that will apply under the directive makes it
inappropriate and impractical to keep contemporaneous records to demonstrate that
the firm has complied with their requirements. There are particular difficulties in
identifying what records requirements would be appropriate to demonstrate
compliance with a rule that specifies very general high level standards of conduct. As
asked in our earlier submission, what records would a firm keep to demonstrate that it
has acted at all times "honestly, fairly and professionally in accordance with the best
interests of its clients" (article 19(1))? At another leve l of detail, there may be
difficulties in identifying the records that can be kept where the rules require a firm
not to do something, as a generalised record keeping obligation would otherwise seem
to require a firm to be able to produce records that prove a negative. For example,
what records would a firm keep to demonstrate that it has "avoided accentuating the
potential benefits of an investment service" in its communications with retail clients
(para 2, Box 7)? The most that a firm would be able to do would be to record the fact
that it had considered the issue.
We consider that further general record-keeping measures under Article 13.6 are
unnecessary. Consistently with its intention not to reverse the burden of proof, CESR
should confirm its intention to delete Box 4 paragraph 4. CESR's implementing
measures should simply specify a minimum range of records that are genuinely
necessary in relation to each substantive requirement of the directive. In considering
its proposals, CESR should evaluate whether its proposed requirements are
proportionate and not unduly burdensome by reference to a cost-benefit analysis.
Such an approach would be consistent with CESR’s first bullet point on page 7, that it
‘does not intend to reverse the burden of proof but, rather, to introduce obligations of
record keeping’. If CESR does, nevertheless, continue to consider that a general
Level 2 provision is appropriate, the approach set out in the third bullet point on page
7 should be modified to refer to ‘audit trails to facilitate regulators’ monitoring of
compliance with the applicable rules’.

4.
Tape recording requirement
We continue to believe that it is disproportionately burdensome to impose tape-recording requirements of the kind proposed. As we stated in our earlier response, there are many firms that do not currently record telephone lines at all (and it should be recalled that one effect of the directive may be to bring many more firms into the scope of regulation). For these, the requirement would impose wholly new costs or require a change of business model. Those firms that do currently record telephone lines do not record all the lines over which orders might (on occasions) be placed and do not normally keep the solid state records for the proposed one year period. The cost of CESR’s proposals for a firm with offices in all the major European financial centres has been estimated at: Initial capital cost of upgraded recording systems: This estimate is based on the following assumptions: storage costs €8 per channel per month; 5000 channels in 6 financial centres; hardware and server costs for upgraded on- line recording system, incorporating centralised solid-state storage, targeted deletions, extensive search criteria, fast search/find/playback, and potential to integrate with other systems. Changes and costs of the kind CESR proposes would need to be justified in a cost-benefit analysis by a strong regulatory or investor protection case. We see no evidence that such a case has been made. What would regulators use the records for, and how often would they use them? CESR would need to provide such evidence of tangible benefits before it could justify pursuing its proposal. Nor do firms completely ban the taking of all orders over the telephone on unrecorded lines, as CESR's proposals would require. Taking orders out of hours or on unrecorded lines might be necessary to meet client's requirements, in cases of urgency, while travelling or because of timezone differences. Our response to CESR’s first consultation gave the example where a client is in contact with a member of the firm's senior management out of hours and wishes to place a stop loss order. CESR's proposals would require the firm to refuse to accept that order at that time. The requirement would thus be inimical to investors’ interests, a fo urth important objection which CESR does not refer to on page 7. CESR’s approach should focus on the purpose of recorded conversations. We consider that their main purpose is to enable quick resolution of disputes, errors and misunderstandings arising in a dealing room environment. Recorded conversations are but one means which contributes to audit trails for transactions. CESR appears to argue that recording of conversations would enable regulators to carry out investigations more efficiently. But it makes no sense to require all orders given or received over the telephone to be recorded when orders taken in person are not. Market participants who intended to abuse investors would simply ensure that they did not give or take orders over a recorded telephone line. There is nothing intrinsic about the doing of deals which means that they are the only matter that a regulator might want to investigate: this emphasises that the reason why telephone conversations of deals are taped is to resolve trading errors and disputes quickly, not as a record for future investigations. Furthermore, outside dealing rooms telephones are widely used for other purposes, including the discussion of sensitive legal issues. A widespread requirement to tape all lines which clients might use to place an order would therefore raise significant issues about confidentiality, legal privilege, and firms’ management of telephone communications. For all the above reasons, a recording requirement is not appropriate outside firms’ dealing rooms. Given the diversity of firms and their practices it remains difficult to give the detailed
cost figures that CESR requests. The cost of retrieving information will also depend
on the nature of the requests for information that regulators make. Nevertheless, if
CESR wishes to continue to proceed with this proposal it will need to consider how to
gather more data - perhaps in a separate study - to support a cost benefit analysis. We
would be pleased to discuss this further with CESR.

5.
Outsourcing of investment services

We continue to believe that there is no need to include provisions transposing CESR
standard 127 on the delegation of investment management as part of the
implementation of MIFID. As CESR points out, the UCITS Directive already
contains provisions on delegation for activities within its scope, which reflect the
special circumstances of UCITS funds. There is no need to add to or duplicate these
provisions in relation to UCITS funds, or to apply those provisions to other types of
investment management relationship for which they may be inappropriate (such as
individual portfolio management). In particular, CESR should bear in mind that firms
provide services to a wide range of clients, including professionals and market
counterparties and in many different circumstances. In some cases, delegation may be
the only practical way of providing the service. It is inappropriate for CESR to seek to
restrict these legitimate activities in the way proposed, merely on the grounds that it
purports to attain parity with a directive which applies to a narrow range of very
different circumstances. It is also inappropriate to extend a CESR standard relating
narrowly to individual portfolio management so that it would apply to any service that
might be outsourced by the many different types of investment firm covered by
MIFID.
We therefore support option (c) as a general approach. We also support the aim of
achieving convergence with other initiatives on outsourcing. Of necessity, this should
involve a less detailed approach to regulation at this stage, bearing in mind the state of
development of those initiatives.
6.
Conflicts of interest and the segregation of areas of business

In general CESR’s proposed approach as set out in its first bullet point is right. Strong
emphasis should be put on the clarity of the objectives of prevention and management
of conflicts. But it is important to recognise that conflicts exist and may be
unavoidable ; that they can be managed, or (as set out in Article 18) disclosed.
Management of conflicts should be permitted in ways that comply with applicable
cultural norms.
However, this approach is not reflected in some of CESR’s other proposals, in
particular CESR’s fourth bullet point (see detailed comment below).
Second bullet point
CESR should remove any possible ambiguity in the second bullet point on page 9 by
making clear that discretion must be introduced as regards the means to manage
conflicts. Article 18 is drafted on the basis of the responsibility of firms to identify
and manage conflicts and the necessary discretionary judgements that this
responsibility entails. Reducing such discretion beyond the determination of ‘criteria for the types of conflicts which may damage the interests of clients’ and ‘steps that investment firms might reasonably be expected to take’ would exceed the permitted scope of Level 2 legislation and undermine the Level 1 Directive. Without such discretion, it would not be possible to deliver the approach set out in CESR’s first bullet point. Effective controls need flexibility, because firms operate a variety of business models and clients have different expectations. Different conflicts arise in different areas of business, with different interests at stake and different degrees of sophistication and different expectations of clients. Flexibility need not be provided at the expense of effective protection; indeed, effective protection arises as a result of clarity about objectives. Third bullet point We agree that the list of examples suggested in the consultation document should not be considered as exhaustive. Fourth bullet point As regards the management of conflicts affecting portfolio management, corporate finance (including underwriting and/or selling in an offering of securities and advising on mergers and acquisitions), and proprietary trading, CESR proposes that ‘information barriers, such as Chinese walls, should not be mandatory, but other means should at least be as effective’. Such a standard would amount practically to a requirement to institute such information barriers, since Chinese wall arrangements are the ultimate structural remedy to manage conflicts of interest, short of a complete spin-off of the relevant areas. As explained in detail in the following paragraphs, we are very concerned that such a rigid requirement would: (a) Unintentionally cover a range of business activities for which it is clearly not (b) Apply too broadly, since it would cover all the conflicts between the relevant business areas and other areas of the firm; (c) Contradict and undermine the principle-based approach of Article 18, which puts the prime responsibility for identifying and managing conflicts on firms; (d) Deprive clients of the sophisticated investment services they demand today from integrated investment firms, and severely limit the ability of European capital markets to innovate.
(a) The proposed requirement would unintentionally cover a range of business
activities for which it is clearly not meant.
Information barriers might be the right means to manage conflicts between some
activities falling within the areas of portfolio management, corporate finance, and
proprietary trading, and some other activities of investment firms. But the activities
as described extend far beyond the activities which we think that CESR meant to
address. We highlight below examples of the kinds of services and business
structures that would be endangered by such a requirement in the three broad business
areas that CESR describes in its paper.
Portfolio Management. It will be appropriate in many circumstances to separate
certain portfolio management activities from certain other services by structural
information barriers, as it is currently often done between asset management
businesses servicing pension funds, managing publicly distributed funds, etc on the
one hand, and brokerage activities on the other. But there is a range of legitimate
businesses which combine portfolio management activities with brokerage or other
financial services. For example, it is a common support model for the management of
the assets of individuals or families, to combine portfolio management, advisory,
brokerage, custody and potentially other services. Private investors will often want to
give orders to the same individuals who are already providing them with advice, and
whom they trust to make discretionary decisions on the management of their
portfolios. How the conflicts of interests potentially affecting such a bundle of
services are best ma naged should not be regulated by rigid principles.
Corporate finance, including underwriting, selling in an offering and merger and
acquisitions advice. It will often be appropriate to separate, for example, merger and
acquisition advisory functions from certain other functions of investment firms such
as brokerage businesses by means of information barriers. But it would be far too
rigid to say that the broad range of functions included in CESR’s description of
corporate finance all need arrangements comparable to Chinese Walls to separate
them from all other activities of a firm. For example, the placement of offerings
would usually be undertaken by staff, who are also engaged in order taking activities
in other contexts. The conflicts which this presents need to be managed by relevant
procedures, which might for example dictate the point of time at which
representatives should be informed about an offering. Management of such conflicts
by arrangements comparable in their scope and effectiveness to Chinese Walls would
be impractical and unnecessary.
Proprietary trading. It would be impractical and inappropriate to apply a rigid “as
effective as” test to the complex management issues raised by the own-account
trading activities of firms across different asset classes. Firms engage in a wide
variety of such activities. The OTC derivatives markets, including commodity
derivatives, and large parts of the fixed income markets rely entirely on the own-
account dealing activities of firms. Separation of client and proprietary trading by a
Chinese wall is not meaningful in the context of a non-exchange-based market. Own
account trading also plays an important part in the equities market and derivatives
trading activities of firms, including hedging activities to manage the overall or
specific risks relating to firms’ market making or derivative portfolios and trading
activities aimed at maintaining appropriate inventory. How the ensuing conflicts will
need to be managed, for example vis a vis agency trading activities for investing
clients, will require complex judgements. In some cases the nature of the conflicts
will require that certain businesses of firms are separated by Chinese walls, but a rigid
rule that all conflicts involving ‘proprietary trading’ activities should effectively be
‘walled off’ would deny investors the benefits that integrated brokerage houses can
bring through liquidity provision. Currently, investment firms’ clients have access to
large pools of liquidity and are able to select their preferred method of execution of a
trade: whether on a pure agency basis, hybrid agency (e.g., the trade is executed in the
market, but the investment firm guarantees a price) or on a principal basis. The
proximity of the principal and client trading functions of an investment firm enhances
the investment firm’s ability to offer competitive pricing for the securities purchased
or sold to the extent that the clients’ demand matches the trading or inventory of the
investment firm. On the other hand, should a client require execution of its business
on exchange, it would be able to instruct the firm to this effect. It does not appear
justifiable to deprive the market of such synergies, especially since the experience of
the European markets shows that fair dealing principles, and order handling rules,
supported by the firm’s policies and compliance checks, can effectively guard the
clients’ interests against any competing interests of the investment firms. Our
Members consider that there is clear support from their clients for the benefits of this
business model, which would effectively be prevented by Chinese walls or similar
procedures.
(b) The proposed requirement would apply too broadly, since it would cover all
the conflicts between the relevant bus iness areas and other areas of the firm.
The proposed ‘at least as effective’ standard does not make clear which conflicts, with
regard to the relevant business areas, would need to be managed according to that
standard. As stated, it would effectively apply to all such conflicts. This would
clearly not be appropriate, since such conflicts will differ in their materiality.
(c) The proposed requirement would contradict and undermine the principle-
based approach of Article 18, which puts the prime responsibility for identifying
and managing conflicts on firms .
The Level 1 text puts the prime responsibility for identifying and managing the
relevant conflicts on firms. Article 18(3)(a) prescribes the scope of Level 2 measures
as being to define the steps that firms might reasonably be expected to take to
identify, prevent, manage, or disclose conflicts. The proposed ‘at least as effective’
requirement, not tied to any particular identified conflicts or the prescribed
reasonableness standard, would, contrary to Article 18, impose prescriptive
requirements in a blanket way on the relevant business areas. Other reasonable steps
for managing conflicts may include internal policies, training, and compliance checks.
(d) The proposed requirement would deprive clients of the sophisticated
investment services they demand today from integrated investment firms, and
severely limit the ability of European capital markets to innovate.
Since the management of conflicts and the decision whether a particular conflict
requires information barriers such as Chinese walls involves difficult judgements
relating to many different businesses, it is hard to predict all the circumstances in
which such a rigid requirement would deny clients the services which are already
customarily provided by integrated firms, or the services it might affect in future.
Many commentators have stressed the need to draft Level 2 measures sufficiently
flexibly to cater for the complexity and diversity of European financial markets, and
to leave room for innovation and future developments. To achieve this objective it is
essential, especially in areas such as conflict management, that a principle-based
approach is adopted, imposing a high level of responsibility on firms to run their
affairs in accordance with stringent high- level standards, subject to regulatory
supervision. The prescriptive approach proposed in CESR’s fourth bullet point would deny firms and regulators the required flexibility in this regard. Finally, CESR’s proposal that “other means should at least be as effective” as Chinese Walls is hardly in line with the mandate to define “the steps that investment firms might reasonably be expected to take to identify, prevent, manage and / or disclose conflicts of interest”. On the one hand, the proposal deprives not only the investment firms but also the home state regulators of any degree of discretion as regards the means by which to manage conflicts. On the other hand, a widespread use of structural information barriers would substantially impinge on the structure of most investment firms, very possibly in an unreasonable and retrograde way. Other reasonable steps include internal policies, training, and compliance checks. Fifth bullet point We take the last statement that “if alternative means were used by investment firms, there should be clarity about this use” to mean that if investment firms decide to use conflicts management tools other than Chinese Walls, this should be made clear, presumably through disclosure to (potential) clients. We would have no objections in principle to such a requirement. However, we should point out that clients are normally aware of the nature of the business of integrated investment firms and of their general structure and do not expect there to be Chinese Walls segregating business areas except for particular areas such as the barriers between investment banking and investment research. To the extent that the level 2 measures will not mandate the use of Chinese Walls in all circumstances no similar expectation would arise and therefore such disclosure would be somewhat redundant. Investment research

Firms need to be able to disclose that their investment research is not impartial if that
is, in fact, the case. If it is clear that a reasonable person would not think of the
research as impartial, the firm should not be forced to attempt to prepare the
research in accordance with requirements relating to impartiality. CESR should
continue to consider the interaction between its proposals and compliance and
disclosure requirements under the Market Abuse Directive, which would apply to
non- impartial research.
We agree with the statement that "there are different situations which deserve
different treatment". We also agree in principle with the statement that "clear
disclosure should be imposed where firms do not fully comply with all requirements".
As regards the latter statement, we trust that CESR would not find it useful to
mandate "itemised" disclosure but would rather recommend generic disclosure.
Whatever the aspects of non-compliance with the IOSCO measures, a piece not so
compliant will equally be "non-objective". Whilst it will be important to point out to
the user that the piece cannot be relied upon as objective, we believe that requiring
investment firms to list each and every specific aspect of non-compliance would not
provide any additional benefit to the user whilst imposing a significant burden on the
investment firms.
Methods and arrangements for reporting financial transactions

In general we welcome CESR’s intention to provide advice to the Commission which
is largely unchanged from that on which it consulted earlier. As we commented then,
this approach is consistent with the major policy issues as we see them. Where the
consensus views of respondents differed from CESR’s original draft advice we also
welcome the positive approach which CESR has adopted in incorporating a number of
those views in its amended advice.
In particular we welcome
The acceptance that reporting in non-electronic format will be acceptable in certain exceptional or temporary circumstances, though non-electronic reporting may be appropriate in a wider range of circumstances than those listed in paragraph 12. The competent authority should have discretion to accept reports in non-electronic format. It will be important to ensure that firms are not obliged to instal electronic systems where the benefit does not justify the cost. The decision to move the issue of service level agreements from Level 2 to Level 3, though we reiterate the concerns expressed in our earlier response about a positive obligation to have a service level agreement at all. The removal of the open-ended nature of minimum condition (g) On the other hand, we hope that CESR might consider further the following issues which we raised previously (a) Amending the draft level 2 advice at (3) to make the granting of a waiver automatic when a firm uses an approved reporting system. In its discussion of possible future work at Level 3, CESR appears to have accepted the validity of the argument. We therefore do not understand CESR’s reluctance to advise on Level 2 accordingly. (b) The problem of monitoring approved reporting systems operating in more than one jurisdiction remains unresolved. Although, as CESR has acknowledged, some respondents argued for a ‘practical’ approach, and CESR’s discussion of the issue indicates a positive sentiment towards resolving the problem, we would suggest that the proposal to work at Level 3 to ensure that competent authorities use the same ‘approach’, while helpful, addresses only half the problem, if it is CESR’s intention that such systems would be subject to ongoing verification of compliance with the general minimum conditions by the releva nt competent authorities. (c) CESR should include in its draft advice the requirement in paragraph 5 on page 12 that the competent authority approve any reporting channel that complies with the general minimum conditions. 9.Criteria for assessing liquidity in order to determine the most relevant market
in terms of liquidity for financial instruments
We support the pragmatic and flexible solutions that CESR has proposed for
instruments for which the proxy approach is problematic. In particular, for
instruments where a proxy is not suitable, we support CESR’s proposal to make only
an initial computation of liquidity, rather than an annual computation, and to base that
on whichever measure, volume or turnover, can be computed with the ‘minimum of
cost and effort’.
In the bond market, the cost element of creating, maintaining and updating a list of
financial instruments as required in paragraph 11 of the draft advice is particularly
significant. Whereas new issues of equity are comparatively rare, whether as IPOs or
secondary, new bond issues may amount to 10 or more a day across Europe. That
number will be substantially larger if daily issuance under medium term note
programmes (MTN), many of which are listed, have to be recorded. Maturing issues
will also have to be removed. We understand that CESR has set up a Technical Task
Force to consider this and other issues and we look forward to CESR’s exposure in
due course of its preferred solution.
We support the recommendation not to disclose publicly the most liquid market for a
particular instrument. Although that will a fact generally known to market
participants, it would be unfortunate if exchanges were able to exploit such regulatory
‘endorsement’ for commercial advantage and to the detriment of competitors.

Draft advice on Cooperation and Exchange of Information related to transaction
reporting (Article 58)
We appreciate the work that CESR has undertaken to develop a solution to the
difficult, but increasingly common, position of remote members of regulated markets
under Article 25. CESR’s proposition, as set out in paragraph 11 of the Explanatory
Text, appears to resolve matters in a way with which competent authorities should be
satisfied. The competent authority of the regulated market receives reports of all
trades on the regulated market; the competent authority of the most liquid market (if
different) receives those reports; and the competent authority of the remote member
also receives those reports. We hope that the indirect nature of the last reporting line
will not prove legally insurmountable since its practical advantages are significant.
We reiterate our previous comments on the need to eliminate duplicative reporting
wherever possible.

10. The minimum content and the common standard or format of the reports to
facilitate its exchange between competent authorities

We welcome the further explanation CESR has provided concerning the issues of
minimum content and the common standard or format. In particular, the flexibility
inherent in paragraphs 11 and 13 of the explanatory text is most helpful.
On the other hand, we are still unclear about the rationale for describing, in paragraph 3, the contents of Annexes A and B as the minimum that competent authorities need for the ‘detection, investigation and enforcement of market abuse, as well as for the other regulatory purposes transaction reports are used for’. Putting to one side the open-ended nature of the ‘other regulatory purposes’ justification, which makes comment difficult, we would suggest that a test of regulatory purpose which better meets a cost benefit test for both transmitters and recipients of transaction information is the minimum necessary to alert a regulator as to the possible need to commence an investigation. If this test were to be adopted, the probable result would be that the amount of information routinely transmitted on every trade would be substantially reduced, with resulting savings in systems, analytical resources, bandwidth etc. CESR states that the minimum requirements set out in Article 25(4) do not include information concerning trading capacity of the firm, the counterparty, etc. It believes that the information required by Article 25(4) is not sufficient to facilitate the use of transaction reports for the detection of market abuse and other regulatory purposes. It is our view that unless a regulator manually reviews each transaction report (which would be impossible in larger markets), automated review must inevitably rely on the information required under Article 25(4). This automated review flags individual transactions for further review by a human being. Further information would be provided at the regulator’s request, but only concerning the transactions that had generated the alert. As the number of transactions that may cause an alert is a very small proportion of all trades, this would be the most cost-effective way of administering this requirement. In some jurisdictions, the proposed minimum fields are additional to the reporting requirements that currently exist. In some Member States transaction reporting obligations are restricted to activity undertaken on a regulated market, so that new reporting systems would need to be developed. These differences will be a major contributor to the system changes that firms need to make before the deadline for implementing the Directive (see 1(c) above and the Annex to this response). We agree with CESR’s decision (explanatory notes, paragraph 14) not to propose advice on a unique Europe-wide client/customer identification. This would be a very costly and burdensome exercise which could not be justified without thorough analysis of the implications and articulation of any benefits, and consideration of the constitutional and human rights aspects in some Member States. As regards the detailed content of Annex A, we wish to make three comments: Trade value/value notation. We welcome the removal of these fields from Annex A for the reasons set out in our previous response. Underlying Instrument Identification. We reiterate our earlier comment that complex structured products may have up to 200 underlying instruments, constantly substituted in and out of the portfolio during the life of the product. For such products this field may therefore be very large and time consuming to populate, while its content will be of very short-term relevance. Trading Capacity. This field description is generally acknowledged to be difficult to define briefly, as CESR recognised in asking for advice in its first consultation. The new language in Annex A is clearer that previously. However, the new language in Annex A is still not entirely consistent with the (largely unchanged, presumably in error: we refer to previous comments on the difficulty of identifying transactions as proprietary) language in Annex B and both carry embedded within them concepts of agency law which may not transpose easily across judicial systems. We may be able to assist if CESR was to explain the regulatory purposes for which this field is to be used. ANNEX: MIFID provisions where transitional measures are most needed
Extra time
Extra costs?
How costs
required:
required
incurred
Documents
implementation
distribution of documents plus staff time spent handling complaints from non-retail clients according to (retail) policy for some firms, entire new recording systems to be purchased integrated nature of large firms’ systems
Description of how a large investment bank would typically go about
commissioning systems changes required by new legislation and/or new regulation.

The implications for system development time scales for MIFID must be put in the
context of the general move towards automation of the whole front-to-back cash
equities order capture, management and execution processes among brokers. The
major broker/dealers are now receiving orders from their clients, routing them
internally to the correct execution desk/location, managing a variety of execution
strategies, passing orders to the exchange, receiving and booking executions and
passing details back to clients with little or no manual intervention for a large
proportion of their flow.
As a result, the systems architecture components are highly interdependent, with a
corresponding need to ensure that data are captured and passed between them
accurately. Project timescales are influenced by, and reflect, the factors below.
1. Clear, unambiguous and detailed requirements are needed before work on the
systems solution can begin in any real sense. The MIFID requirements have
to be fully understood by project teams (that is, involving many personnel who
will have had little or no previous involvement with the content of MIFID).
One would anticipate an initial set of meetings with in- house MIFID experts,
with subsequent meetings, if necessary, with the appropriate regulatory body
to clarify ambiguous points. The time needed is likely to be extended by the
fact that the whole market will be seeking similar meetings.
Approximate time from publication of final proposals: 2 months
2. Thorough analysis of all systems in the front-to-back flow will be required to
ensure that the changes required to each component are identified and
specified. This becomes more crucial where changes to trade or transaction
reporting are required, as the appropriate data has to be captured at the
appropriate point in the process and passed between systems to ensure
accurate and timely reporting results.
Approximate time taken:
3-6 months
3. Develop full testing plan – this work will be done concurrently with the other
4. After the coding of any new systems or changes to existing systems, the
interdependence described above makes it necessary to establish a testing
environment covering all systems in the flow and to develop a test plan and
test data that are suitable for testing the changes and ensuring that there is no
unintended impact on existing functionality. A number of test cycles, with
allowance for retesting of corrected bugs, will need to be planned.
Approximate times: Coding
2-3 months
Setting up test environment
Internal test
Coding corrections
2nd test period and adjustments;
test to clients
Final test
5. This becomes more complicated with the inclusion of exchanges, where the availability of test environments varies from exchange to excha nge and is an
external constraint to brokers. This complication will be compounded by the
fact that the whole market will require access to the same test environments.
Historically, exchanges have sometimes had to schedule dedicated time slots
to each broker, which introduces additional time lags between testing and
retesting for individual brokers.
Approximate time:
2-4 months
It is also possible that exchanges themselves will require brokers to participate
in market-wide testing of the exchange system. This will require additional
time.
Approximate time:
6. Many brokers depend on third party systems, where several brokers will be contending for a finite number of development resources capable of integrating and testing the changes in their overall environments. 7. The above analysis suggests that the total time needed from the pub lication of final regulatory requirements to implement such system changes is in the region of 18 to 24 months.

Source: http://www.vpff.no/content/download/1344/8126/version/1/file/Joint+letter+to+CESR+regarding+MIFID+17.12.2004.pdf