AMTE's Response to the European Commission's Call for Evidence by mmcsx

VIEWS: 5 PAGES: 60

									    AMTE’s Response to the European Commission’s
                 Call for Evidence

« Pre- and post-trade transparency provisions of the markets in Financial
  Instruments Directive (MiFID) in relation to transactions in classes of
                financial instruments other than shares »




                                                          September 2006
Table of contents



1. AMTE and the AMTE Working Group on Bond Market Transparency in Europe.. 3
2. AMTE’s View on Regulation............................................................................................ 4
3. Summary and Recommendations .................................................................................... 7
4. Retail Participation in European Bond Markets.......................................................... 10
5. Transparency vs. Liquidity............................................................................................. 12
6. Equities vs. Bonds ............................................................................................................ 16
7. Availability of and Payment for Post-Trade Data ........................................................ 17
8. Repo Markets and Liquidity Portfolios of Banks and Insurance Companies ........... 18
9. Macro (or Systemic) Transparency vs. Micro Transparency ..................................... 19
10. Primary Bond Markets Transparency......................................................................... 21



Appendix 1: Key Points Arising from AMTE Response to FSA Discussion Paper 05/05 on
Trading Transparency in the UK Secondary Bond Market...................................................... 23
Appendix 2: AMTE Response to FSA Discussion Paper 05/05 on Trading Transparency in
the UK Secondary Bond Market .............................................................................................. 26
Appendix 3: Empirical Evidence on the Daily Returns Distributions in the European Bond
Markets..................................................................................................................................... 37
Appendix 4: Response to Specific Questions Raised by the European Commission ............. 47
Appendix 5: AMTE Working Group “Bond Market Transparency in Europe” Members..... 56
Appendix 6: AMTE Board members and list of AMTE institution members........................ 58




                                                                                                                                             2
1. AMTE and the AMTE Working Group on Bond Market Transparency
   in Europe

AMTE (Association des Marchés de Taux en Euro), the Euro Debt Market Association, was
created in July 2002 and counts now 61 members, leaders in their field and representative of
the euro debt markets: issuers, financial intermediaries, investors and other market
professionals.

The association aims to facilitate the increase of the depth, liquidity, transparency and
innovative nature of the euro debt markets. In this context, AMTE’s purpose is to contribute
to the development and effective operation of euro fixed income and derivatives markets by
supporting the various professionals in these markets in their efforts to come to an agreement
on concrete ways to reach these goals.

To achieve its objectives, AMTE conducts work and research, and organises consultations
among its members on all subjects relating to any type of operation on euro-denominated
financial markets excluding equity markets.

In November 2005, AMTE established a Working Group dedicated to bond market
transparency in Europe with a view to providing input in the context of the review of Art. 65
of the Directive 2004/39/EC on markets in financial instruments (MiFID). In response to the
Financial Services Authority’s Discussion Paper 05/05 on trading transparency in the UK
secondary bond markets, AMTE issued a Thematic Paper, a summary of which is attached
hereto in Appendix 1. The full text of the Thematic Paper is also attached hereto in Appendix
2.

The Working Group has now been reformed in order to develop a response to the call for
evidence on non-equity market transparency by the European Commission (the
“Commission”).        This Working Group still comprises representatives of the three
constituencies of the bond market (issuers, banks and investors). The scope of the Working
Group initially focused on vanilla bonds at the first stage, however, given the scope of the
Commision’s call for evidence, we have felt it necessary to widen that scope to include
interest rate swap (IRS) and credit default swap (CDS) markets, particularly given their
importance in pricing bonds. This Working Group will likely be in existence for some time
as the MiFID process for bonds and other financial instruments process will take a number of
years before implementation. The Working Group will remain available to respond as AMTE
sees fit to the ongoing debate on MiFID for non-equity financial instruments.

While this entire paper together with its appendices is submitted in response to the
Commission’s call for evidence, summary responses to the Commission’s specific questions
are contained in Appendix 4.




                                                                                            3
2. AMTE’s View on Regulation

It is AMTE’s opinion that the health and attractiveness of the European bond markets depend
on the following key factors:

               1. The financial integrity of the participants and of the market as a whole;
               2. Market driven liquidity;
               3. Integration, efficiency and robustness of the clearing and settlement
                  system;
               4. Efficient price formation and trade execution;
               5. Minimisation of asset price bubbles.

Europe exists in a competitive world and the internal benefits generated by fair and efficient
financial markets can be augmented if market players outside Europe are encouraged to use
the European market place to issue, invest and trade financial instruments, not just in
European currencies but in all the currencies of the world. Against this must be considered
the impact of inadequate or inappropriate regulation driving the market out of Europe. In the
current context, and certainly until the current MiFID regulations have been implemented and
tested in practice for some time, it remains to be seen where the balance of regulation lies.
The historical shift of market activity from New York to London as a result of Interest
Equalisation Tax and the current concern about a possible shift of private banking business to
Singapore as a consequence of Directive 2003/48/EC on taxation of savings income in the
form of interest payments are two examples of the possible consequences of regulation. It is
granted that both these measures are tax measures, but the market reaction illustrates the
ability of markets to migrate. Another more recent example unrelated to taxes, is the
increasing preference of, in particular, foreign bond issuers to list their bonds on unregulated
rather than regulated markets within the European Union (EU) or the European Economic
Area (EEA) or outside of the EU and EEA as a reaction to the new regime imposed by
Directive 2003/71/EC on the prospectus to be published when securities are offered to the
public or admitted to trading (the “Prospectus Directive”).

The last example shows that, just as market activity can shift geographically, so can it migrate
to other less regulated markets. Regulation designed to protect retail investors in the cash
bond markets may have the effect of driving liquidity to less regulated and less transparent
sectors. The number of corporate issuers who have issued in minimum denominations of
EUR 50,000 in order to simplify the prospectus requirements of the Prospectus Directive is a
concern to the extent that it restricts retail choice. In addition, the costs of regulation by
increasing barriers to entry may lead to less competitive markets.

Against this background, we agree with the two main policy objectives outlined in the
Commission’s call for evidence; namely consumer (in particular retail investors) protection
and market efficiency while cautioning against mandatory measures to achieve these
objectives unless there is clear evidence a) of market failure; b) that regulatory intervention


                                                                                              4
will improve the situation, and; c) that the benefits brought by regulatory intervention will
exceed the costs having taken into account the possibility of unintended consequences of
regulation.

Consumer protection can be a justification for regulatory intervention, and the Commission
rightly places retail markets at the top of its priority list. Fair and transparent pricing,
allowing consumers to compare prices across a range of products, is normally a sine qua non
for adequate consumer protection. It is questionable whether this is achievable in practice for
retail investors across the totality of the bond markets given the vast range of bond issues on
offer 1 . Price comparisons, however, can be made using swap and government yield curves
and the still relatively intransparent CDS market 2 . This is beyond the capability of all but a
very few retail investors and no doubt a number of professional investors. A number of single
and multi dealer business to customer electronic platforms, however, allow such comparisons
on a range of products. The pace of market development is such that it is probable that
wealthier and larger retail investors will ultimately be provided with quite sophisticated tools
for pricing bonds. In addition, we are likely to see sophisticated web based pricing tools
available to all classes of investors. Ultimately, sophisticated individuals will be able to price
even the most complex structured products known today.

In the case of consumer protection it must nevertheless be clear that the benefits to the
consumer outweigh the costs which will be passed on, but more importantly that consumer
choice is not thereby restricted and consumers offered more expensive and less appropriate
products. Most retail savings products are substantially less transparent than fixed income
bond investments; that retail investors none the less invest in less transparent savings products
may say more about the relative commissions available to intermediaries that any lack of
transparency in the bond market.

We believe that the Commission through its public pronouncements and many of the
implementing measures of MiFID have taken these points on board but they none the less
bear repeating.

We do not accept as an adequate argument in favour of regulation simply that technological
developments make it possible. Our general view is that increased transparency resulting
from technological developments has made it possible to regulate less.

The Commission’s call for evidence asks for evidence that the European secondary non-
equity markets operate in a sub-optimal way. We question whether the mere fact that markets
are sub-optimal is sufficient to justify regulatory intervention, though, as is suggested by the
Commission, self-regulation might be a suitable alternative. No market in practice fits the
description of a perfect market. A market which was optimal in all ways would be a static

1
 There are 52,000 non-government bonds issued in the European markets (source: Bloomberg).
2
 Fixed income markets are more homogeneous than equity markets and most fixed income instruments will
move in the same direction as a result of macro economic events. The major exception, of course, are bonds
subject to credit events.



                                                                                                             5
market. Europe needs dynamic markets adjusting as a result of technological innovation and
competitive forces. As has been argued in the papers referenced below (those by the BMA,
CEPS, CEPR and the UK FSA in its Discussion Paper 05/5 and Feedback Statement 06/4),
competitive forces and new technology have brought very significant improvements to
European bond markets in recent years and will continue to do so, particularly in the market
for euro denominated debt.

The UK FSA in its excellent work on bond market transparency has set a criterion of
“market failure” 3 to justify regulatory intervention. We feel that this standard of
“market failure” represents a better benchmark against which to justify regulatory
intervention. As we argue in this paper, “market failure” is more significant at the macro
level where it is evidenced by occasional market instability or a major dislocation of market
liquidity. It could also exist in less obvious form where market wide historical information is
collected by regulators but not made available to the market for the purpose of analysis and
proper evaluation of market value and performance as well as of best execution. We do not
see any significant instances of the indicators of “market failure” provided by the FSA,
except to some extent in the retail market. There is some concern amongst issuers about a
possible withdrawal of investment bank capital resources from cash bond market-making
which could result in an increased cost of funding for issuers due to investors requiring an
illiquidity premium.

We are particularly concerned that regulatory intervention is, by necessity, relatively
inflexible and designed around a model of the market existing at a point in time (or worse still
an academic model that does not exist in practice). Europe’s bond markets are highly
dynamic and changing rapidly. Attempts to build a mandatory transparency regime are likely
to hinder further market development. To the extent that levels of transparency are felt to
significantly hinder market efficiency and consumer protection, we strongly favour a dialogue
by regulators with market participants resulting in self-regulation rather than immediate and
inflexible mandatory regulation. There is no reason why the Commission should not maintain
a watching brief over developments and if conditions do not improve sufficiently by self

3
  In its discussion paper DP 05/05, the FSA considers that “a market failure might manifest itself if:
• there might be inefficiency in the price formation process, with some participants being unable to judge at
what price to place orders and whether it was appropriate to hit/lift quotes; and
• that there might be a failure of best execution, if those who owed a duty of best execution failed to find the best
obtainable prices for their clients.
Further, such failures might be reflected in:
• wide spreads might indicate insufficient competition (although we [the FSA] recognised in the DP that
determining the size of spread in a given bond that would indicate a market inefficiency would be very difficult);
• a wide dispersion of prices for very similar trades in the same instrument and around the same time (this might
reflect an informational inefficiency, with some participants materially better placed than others to judge at what
prices to trade);
• low participation rates might reflect a lack of confidence on the part of some types of investor to trade in these
markets, or that some form of barrier to entry might exist; and
• a high level of user complaints might reflect that a market failure of some form existed.”




                                                                                                                   6
regulation, and further research indicates a clear net benefit from regulation, then regulatory
action can be taken.


3. Summary and Recommendations

In this paper we consider if and how key factor 4. above (efficient price formation and a
reputation for fair execution) can be met by improved secondary, as well as primary, market
price transparency (which we refer to as micro transparency) while maximising market
liquidity by ensuring that dealers are incentivised to commit capital to trading financial
instruments by obtaining adequate returns on the capital committed to the market. The trade-
off between transparency and liquidity is addressed in some detail in the body of this report.
At the same time, and following the lead of the European Central Bank (ECB), we consider
issues of macro (or systemic) transparency and, in particular, how measurement and
publication of information on aggregate market positions 4 could lead to a reduction in asset
price bubbles. It would be a strange world if regulation at a European level focussed on
ensuring that the smallest participant obtained the best price on a small bond transaction,
while at the same time ignoring the major damage which could be inflicted on small investors
and pension funds by the existence of large asset price bubbles. This latter factor has knock
on implications for the financial integrity and stability of the markets, particularly through the
market for secured lending (the repo market). In this respect, we consider how improved post-
trade transparency could improve the functioning of the repo market without impacting on the
liquidity of the cash market. Secured lenders are not the only actors outside the immediate
arena which would benefit from improved post-trade transparency; risk managers, boards of
directors and trustees, regulators of pension funds, bank liquidity portfolios and insurance
companies, rating agencies and others would all benefit from improved post-trade
transparency but more importantly would benefit from better indicators of systemic risk
resulting from improved systemic transparency in the market.

Post-trade transparency is beneficial for monitoring best execution. No review of the
transparency of European secondary bond markets can be undertaken without taking into
account the potential impact of MiFID implementation, in particular with respect to best
execution rules, and the continuing impact of technological innovation in secondary bond and
other non-equity market trading. These factors have or will improve consumer protection and
transparency significantly.

Most academic work on markets is based on the broker market model rather than the dealer
market model. Fixed income markets naturally fall into the dealer market category because of
the greater ability of, or indeed need for, dealers to hedge and hold inventory. This improves
liquidity of the market and reduces the cost of capital for issuers. In equity markets dealers
are less able to hold inventory and have less satisfactory hedging instruments thus the market

4
  By this we mean positions which might have a significant distorting effect on prices or result in systemic risk to
the market.



                                                                                                                  7
gravitates to the broker model where maximum pre-trade transparency and the use of
regulated exchanges is necessary in order to match buyers and sellers simultaneously.

One further area where bond markets differ from equity markets is the ability to price bonds
using IRS and CDS. In particular, the more recent CDS market is less transparent and we
shall consider whether this calls for some form of regulatory intervention. It should be noted
here that transparency is usually a function of the age of a market and in the early days of a
market transparency will usually be low, allowing expert market participants a considerable
advantage over non-expert participants. This is reasonable as the risks in a new market are
high and the potential for above-average profits are necessary to attract new players into the
market. Many new markets and products fail to take off leaving sponsoring dealers with a
large capital investment to be written off and possibly unsellable financial products on the
books. Furthermore, product standardisation is typically low in new markets. With the
product maturing over time standardisation (or commoditisation) of and, simultaneously,
liquidity in the product increase which in turn results in increased levels of transparency due
to higher dealer competition in commoditised markets. In summary, once a market begins to
be established, competitive forces should rapidly bring increased transparency, but regulation
is unlikely to play a positive role in expanding the market or adding to its transparency in
these early stages.

This development was well illustrated in the IRS market in the late 1980’s with the
establishment of Intercapital Brokers (ICAP) (as well as others) as a specialist inter-bank
derivative broker. ICAP sought to maximize its turnover, and hence brokerage commission,
by aggressively seeking to narrow the published bid/offer spread for IRS. ICAP is now the
largest inter-bank money and derivatives broker in the world. Another salient comparison
with the development of the IRS market is that regulators came late to the IRS market. The
Bank of England/Federal Reserve paper proposing regulatory capital on IRS was only
published in 1986 well after the first transactions were concluded in the late 1970’s and the
traded market became established round about 1984. The first public misselling scandals in
the IRS market did not occur till around 1993 5 . Most derivative misselling has involved
complex structures designed to circumvent internal and external rules and regulations. In
many cases products were designed to deliberately hide the degree of risk involved in the
transaction. These had nothing to do with a lack of price transparency.

Overall, the CDS market today, as was the case with the IRS market, have added considerably
to market transparency and dramatically improved price formation in the bond market. They
have done this without regulatory intervention.

We have reviewed the various studies of bond market transparency published in recent times,
in particular those by the BMA, CEPS, CEPR and the UK FSA in its Discussion Paper 05/5
and Feedback Statement 06/4. We leave their arguments to stand on their own. Our


5
    Gibson Greeting Cards



                                                                                             8
deliberations are those of market practitioners and our broad conclusions and
recommendations are as follows:

          1. There is no evidence of market failure in the wholesale bond market, and in the
             retail bond market only to the extent that retail investors do not have easy
             access to the trade data available in the wholesale market at a reasonable cost.
             The balance of benefit to the retail investor, however, probably lies in the
             health of the wholesale markets. Unintended consequences of regulatory
             initiatives affecting the wholesale market might in fact reduce the protection
             provided by efficient and transparent wholesale markets to retail investors.
          2. That mandated pre-trade transparency is increasingly unnecessary in the
             European bond market and presents problems of implementation and probable
             unintended consequences that make it unlikely that benefits would outweigh
             costs.
          3. That there is a sufficient probability that mandated real time post-trade
             transparency in the European bond market would lead to a reduction in market
             maker liquidity, at least for less liquid and less highly rated issues. In addition,
             real-time post-trade transparency might be counter productive for maintaining
             the existing level of commercially driven pre-trade transparency in the
             European bond market.
          4. The Commission should consult with the industry on how increased self-
             regulation could be achieved and maintain a watching brief on developments.
             Consideration could also be given to a more formal system of end of day price
             fixings for bonds - again operating on a self-regulatory basis. Consolidation
             and dissemination of trade data which is currently accessible for wholesale
             investors only and making it available to retail investors at a reasonable cost is
             the key concern here. ICMA TRAX or similar existing reporting or
             information systems provided by market participants may be an avenue to
             address this issue.
          5. In order to address concerns about the transparency of the derivatives markets,
             we believe the Commission should consult with the industry with a view to
             establishing and publishing, on a self-regulatory basis, a series on fixings of
             the more liquid and commoditised derivative instruments including a range of
             CDS as well as for the repo market.

We further conclude that:

          6. There is a strong case for the estimation and publication of aggregate market
             positions and other risk/return measures covering all categories of financial
             instruments. This would require cooperation from all European market
             regulators working in conjunction with market players and trade bodies to
             establish measures of aggregate market positions. The absence of published
             market wide indicators of aggregate positions is in our opinion the largest
             single failure in the European financial markets.



                                                                                               9
One of the paradoxes of the bond markets is that the IRS, the CDS and the repo markets are
the major contributors to liquidity and, by allowing comparison between instruments, to
transparency in the bond market; yet they remain the least transparent of markets themselves.
They are, so to speak, the engine room of the financial markets. Most ships’ captains are not
concerned about the details of what goes on in the engine room, but they do want information
on the output of the engines. Periodic published fixings of IRS, CDS and repo markets would
be of immense value in improving transparency.


4. Retail Participation in European Bond Markets

The vast bulk of retail investment in the bond market is indirect through funds and insurance
policies. While we recognise the importance of direct retail investment in maintaining
consumer choice, the balance of benefit to the consumer probably lies in the health of the
wholesale markets.

We do not argue with the Commission’s focus on consumer protection. This must form the
key rationale for regulatory action. There is much debate about the significance of retail
investment in the bond markets. Direct retail involvement in bond markets varies
significantly between countries as it does between equities and bonds. We accept that some
form of best execution or transparency regime will improve the likelihood of retail use of the
bond market, though whether it is sufficient to significantly increase retail penetration in, say,
France and the UK is a subject of debate. Public debt offices and other major issuers in these
markets have worked hard to improve retail penetration with relatively little success. We
certainly believe that implementation of existing MiFID requirements for transparency of
commissions; best execution and suitability should improve the environment for retail
investors.

We see two major protections for retail participants. The first is the best execution
provisions of MiFID. These will ensure that retail will at least obtain a “fair” price and
implementation of these provisions will require dealers to improve levels of pre-trade
transparency. It is possible however that this will lead to a two tier market with the existing
OTC market where participants ask for a quote (usually a bid or offer rather than both) being
confined to market counterparties and a wider (bid/offer) screen-based and automated market
where trades are on a best execution basis 6 . The second protection for retail is access to
bonds traded in the wholesale market where a price has already been established by
professional counterparties. This is where we see the greatest danger of targeted regulation
working against the interests of retail investors and the unintended consequences of the
Prospectus Directive are salutary7 .
6
  It is likely that post trade market information will be required to check whether trading venues are providing
adequate prices.
7
  The Prospectus Directive mandates a higher level of prospectus disclosure for issues with denominations below
EUR 50,000 leading a number of issuers to elect to issue in denominations of or above EUR 50,000 or to migrate
from regulated to unregulated markets.



                                                                                                             10
We also warn strongly that regulation designed to protect retail can have an adverse effect. In
spite of strong consumer protection regulation in the UK, retail investors have suffered a
series of misselling scandals including but not restricted to, personal pensions; mortgage
endowment misselling; split capital investment trusts and precipice bonds. Simultaneously, it
is reported that it has been difficult for UK retail investors to buy high quality bonds through
their brokers or banks. If this is true, it is a very perverse result given the type of products
which have been sold over the years. Commission structures no doubt play an important part
in this and transparency of commissions charged to retail and above all received by
intermediaries from product providers is probably the single most important component
of consumer protection. As was pointed out in our previous Thematic Paper attached hereto
in Appendix 2, no amount of price transparency would have protected investors in Parmalat
and Argentinean bonds from losses resulting from these issuers defaulting on their
outstanding debt. On the other hand, it may be that more macro information about debt
issuance levels of these two issuers as well as information concerning positions by trading
venues might have been of some use.

A regime that imposes transparency and best execution requirements on high quality
and liquid bonds but not on other products, may well further encourage retail
intermediaries to sell only those other products. The imposition of suitability tests on
these other products may well be insufficient protection. The mere fact that a broker has
asked his client to fill in a questionnaire and the client has, in the hope of obtaining higher
returns, indicated a willingness and ability to buy structured products does not, in our view,
protect the consumer from inappropriate selling.

Increased retail participation in the bond markets is likely to improve the quality of those
markets. This is because a diversity of investors with different investment rationales and time
horizons improve liquidity and price formation. This is notwithstanding the fact that retail
investors normally hold bonds to maturity. This latter factor may be partly caused by the
relatively high costs for retail in buying and selling bonds and also by the relative illiquidity
of most retail targeted bond issues. Italian regulators have done much to ensure that retail
targeted issues sold in Italy are liquid. This has been in response to the past inability of retail
investors to sell structured bond issues which have been popular in Italy. While we strongly
favour a pan European model and relatively light levels of regulation, it may be that useful
lessons can be learned from the Italian experience.

Taken in the round, we see considerable benefits for retail investors and for the market as a
whole of increased retail investor involvement. Diversification of investor bases improves
overall liquidity and our anecdotal evidence is that retail investors can often be highly rational
in their investment decisions with a tendency to “buy low and sell high”. Improved access to
the more liquid wholesale markets will improve retail investors’ ability to sell bonds as well
as to buy them.




                                                                                                11
There is none the less considerable scope for greater financial education of individual
investors and perhaps trade associations and individual market participants could have a
greater role here.

In the debate on direct retail investor participation in the bond markets far too much emphasis
has been placed on retail access to primary markets. To a large extent, this emphasis is the
result of equity new issues in connection with privatisation efforts and initial public offerings
(IPOs) which have been priced too cheaply in the equity markets, but it also results from the
historical mechanics of retail distribution. Banks have focussed retail sales efforts on new
issues, because the costs of manually circulating issue details to a wide distribution network
could only be justified for primary placements where a subscription price can be maintained
for a period of time and which includes a relatively large commission, comparable to those
available in the equity new issue markets. Electronic trading platforms, while they do not
exclude retail participation in primary issues, make execution in the secondary market, (where
a fair market price has already been established) a far more economic solution. Mispricing of
primary issues to encourage retail or to provide excess underwriting profits (for underwriters
and favoured clients) is hardly to be encouraged and recent evidence is that privatisation new
issues and IPO’s are increasingly being conducted through professional market tenders or
auctions.


5. Transparency vs. Liquidity

Much of the debate over bond market transparency has revolved around the trade off between
transparency and liquidity. The argument is that market makers are willing to commit the
capital which provides liquidity in part because of the access to information that market
making provides. But, if the market becomes too transparent, dealers suffer in that other
market participants are aware of their positions. In a post-MiFID environment, and to a large
extent today, the customer will be able to compare a single quote with that quoted for smaller
size on a variety of trading venues. In a world of post-trade transparency the customer will be
able to look at the historical performance of its security against various benchmarks and again
estimate the fair price. The problem is that if the trade is reported to the market the market
maker, far from gaining access to exclusive information, will suffer the winner’s curse. Even
if the trade size is not reported, any deviation from the market price for a trade marked as
EUR5millon+ will be seen by the market as evidence that a large trade has occurred. In
consequence, so the argument goes, post-trade transparency will lead to a reduction in capital
which market makers are willing to commit to the market, particularly for lower rated and less
frequently traded bonds 8 . A number of empirical studies have been carried out on the impact
of post-trade transparency with often conflicting conclusions. One common feature of these
studies is that none of them have been carried out using European bond market data for the


8
 Arguably, this is already the case where customers have access to iBoxx data: at any stage any constituent of
iBoxx is given a “fair” price. The iBoxx universe is only the more liquid subset of the bond universe.



                                                                                                                 12
simple reason that adequate data is not publicly available9 . Studies have been based on equity
markets and/or US bond markets. This alone should be an argument for improved post-trade
transparency in the European bond markets and we shall return to this point in section 7.

Market makers will seek to protect themselves against sudden market price jumps by having
the most up-to-date market information and better analysis than other market participants - it
is no accident that market makers on the New York Stock Exchange are known as “experts”.
If they are official market makers they may have privileged access to the issuer but
increasingly transparency rules and quicker dissemination of information are eroding this
privileged status. We think it is not unlikely that beyond a certain point in the evolution of
markets increased transparency will lead to wider bid/offer spreads. What we do argue is that
this increased price transparency should be and has been largely market driven. Regulatory
initiatives forcing transparency has to be weighed against causing liquidity to move from
regulated markets to less regulated markets (either within the EU or outside). The growth of
unregulated hedge funds and private equity funds may be caused in part by excessive
regulation, though this is not to deny that they have performed a useful function in the
markets. Such funds can afford to obtain their information from research organisations that
do not publish their information generally. Currently, there is evidence of a shift in liquidity
from relatively highly regulated cash bond markets to less regulated CDS markets, but this
may be due as much to increased investor demand and a search for yield enhancement.

It has to be recognised that there is no such thing as an objective exogenous market price.
Prices are those quoted by dealers to trade in certain circumstances based on the dealer’s
belief that he will be able to earn a profit on the trade. It is worth considering why and how
market makers need to provide capital to a market making activity. In essence, market
makers are option writers 10 . In return for giving others the option to hit their bids or lift their
offers they hope to capture the bid/offer spread and have access to a privileged flow of
information about market activity. Their risk is that short term price variations increase and
that consequently tight bid/offer spreads dictated by market competition no longer
compensate for volatility and also, more importantly, that large sudden shifts in market prices
leave market makers with significant losses on inventory. The ability of the dealer to earn a
profit, and only this, justifies the dealer in committing his capital to the business.

There are a number of reasons why a dealer will seek to reduce transparency to protect his
trading position:

                 1. To enable him to obtain further information before committing to his price
                    and thus reducing his risk of loss;
                 2. He does not want competitors to piggy back off his prices;
                 3. He can engage in selective pricing.

9
  The UK FSA in its Feedback Statement 06/04 provides a study of bid/offer spreads for European corporate and
government bonds based on data for selected issues provided to it by ICMA. This is an exception that proves the
rule!
10
   See Appendix 3 for further discussion on this point.



                                                                                                            13
Against this, the dealer needs to be able to attract business for which he needs to quote prices.
In practice, he will often publish a slightly wider price than he might otherwise be prepared to
deal on and often quote inside this price to customers who approach him with a specific trade.
The price he quotes to the customer will depend on a number of factors including whether the
customer has a reputation for trading (is a quasi dealer) and the strength of the business
relationship. The information content of the published price is thereby reduced. A dealer
who has a long position will try to avoid lowering his published offer and thereby signalling
his position to the market. He will hope that a customer will approach him asking for an offer
or will seek out a customer who he knows has an interest in buying stock. If however other
dealers start to reduce their offers the dealer will realise that he has to compete if he is to
offload his stock. Thus pre-trade there is a delicate balance between a need to advertise prices
and a desire to disguise them.

The situation is less ambiguous for post trade-transparency. No dealer will want to advertise
his position or the price at which he has concluded trades. The more time that passes after the
trade, the less the dealer’s capital is put at risk if the information is made public. The trade
information is none the less highly useful for a number of actors and has great public value. It
is unfortunate then if the information is not published simply because the dealer’s position
could be exposed. In fact, the post-trade information does have a market value and data
vendors obtain significant income from the sale of data. By delaying the publication of data
possibly till the end of the day, a balance is struck between the possible negative value to the
dealer - which declines rapidly - and the public value of the information - which declines
much less rapidly.

Market-maker prices are not the only prices in the market. Investors will have price levels at
which they wish to buy and sell but in general they keep these secret. No one has suggested,
incidentally, that investors should be required to publish these prices. If, however, market
maker bid/offer spreads become wide, investors may be prepared to step into the market and
publish prices at which they are prepared to trade in the hope of finding the other side of the
bargain. Investors with a significant interest in a market segment may, with the aim of
encouraging liquidity as well as earning a profit, sometimes quote prices to market makers.
These quasi- (or part-time) market makers will be encouraged to enter the market when the
bid offer spread becomes too wide. Issuers, in particular frequent issuers, can also assist
market-maker liquidity by communicating levels at which they will tap issues to their
underwriting dealers. This encourages dealers to provide an offer to the market even when
the market is short of stock. The liquidity of repo markets to allow dealers to cover shorts is
important in this respect, too.

Issuers can also encourage liquidity by awarding new issue mandates to dealers with a greater
willingness to provide liquidity by committing capital for making markets and to trade on
tight bid/offer spreads. Issuers and investors can also benefit by talking to each other. The
possibility of investors and issuers bypassing dealers is remote, however, particularly in
today’s regulatory and competitive environment. There is some merit in investors providing




                                                                                              14
feed back to issuers on dealers’ liquidity provision. The better liquidity providers will not
object at least.

It is difficult at this stage to estimate the impact of MiFID best execution requirements on
transparency. At the very least, we believe that it will lead to an increased use of regulated
exchanges and multilateral trading facilties (MTFs) for best execution purposes and thus
reinforce the strong growth in pre-trade transparency that has been evidenced in recent years.
There is also likely to be a move for single dealer systems to amalgamate so as to better
satisfy best execution requirements. Agency broking may enjoy a resurgence. At the present
time a whole range of dealable quotes can be accessed on Bloomberg page ALLQ. This
argues for a delay, at least, in considering pre-trade transparency regulation until after the
impact of MiFID best execution rules have become clear. Market makers will find it harder
to execute on a best execution basis unless they publish prices at which they are willing to
trade. We do not see any further need for pre-trade transparency than that effectively
mandated for best execution purposes and that which is naturally developing through
technological development.

The argument for and against mandatory post-trade transparency requirements is more finely
balanced. In the first place there is a clear lack of publicly available post-trade data on the
European bond markets, though iBoxx data is significantly reducing that gap for the more
liquid bonds 11 . Post-trade data is necessary to check best execution; to provide portfolio
valuations, preferably on an end of day basis; to value collateral; to provide data, such as
historical volatility for risk managers; to provide estimates of liquidity for insurance company
portfolios; to provide the basis for academic studies: the list is virtually endless. Against this
are the problems highlighted of discouraging market makers from quoting for large orders and
even for relatively small orders in illiquid bonds. To balance these two offsetting factors we
believe that an end of day post-trade reporting regime for all trades in cash instruments on a
self-regulated basis could be a viable solution. ICMA TRAX or similar existing reporting or
information systems provided by market participants may be an avenue to address the
problem of consolidation and dissemination of trade data which is currently accessible for
wholesale investors only and making it available to retail investors at a reasonable cost. OTC
derivatives present a clear problem in reporting and publication given that there are an infinite
range of structures and we believe there would be no useful purpose in mandating such
reporting/publication other than for the purposes of estimating aggregate market positions.
Rather, regular fixings of standardised liquid OTC derivatives transactions and maturities as
well as repo transactions could be published by relevant trade bodies. We feel that self-
regulation in this respect is likely to produce better results than mandated transparency not
least because as the market develops so will the nature of instruments and the manner of
reporting.

Little discussion has taken place about the form that post-trade transparency should take.
Should simple cash prices be used or should spreads to government benchmarks and or

11
     It should be noted that iBoxx prices are not necessarily dealable prices.



                                                                                               15
interest rate swaps be published. The only price that is certain is the cash price (and yield to
maturity) paid for a bond (except for transactions concluded at par on an asset swap basis 12 ).
Spreads to government bonds and to swap curves can be rather more subjective though
increasingly bargains are struck on this basis. This is a difficult issue as spread information is
usually the most useful, at least to professional investors, though retail investors are likely to
benefit from the information as well. The alternative to publication of spread information
would be to use published fixings to generate spreads from prices. This would cause a
problem for trades executed away from fixing times in volatile markets. We do not propose a
solution to these problems at the present, merely to highlight them.


6. Equities vs. Bonds

There is no doubting that there are fundamental differences in the nature of equity and bond
markets that have major consequences for liquidity and price transparency. For regulators to
try to impose a one size fits all regime to the two markets would be akin to trying to put a
square peg in a round hole.

According to Bloomberg there are around 52,000 non-government bonds outstanding issued
by European issuers with an average nominal value of EUR 130 million and a total value of
EUR 6.7 trillion. The top ten non-government issuers have 28,700 issues outstanding with an
average size of EUR 60 million and a total value of EUR 1.7 trillion. The top ten EU
government issuers have 2,200 issues outstanding with an average size of EUR 1.6 billion and
a total value of EUR 3.5 trillion. Against this there are 6,555 European equity issues listed on
European stock exchanges with an average market capitalisation of EUR 1.6 billion.

Bonds usually have a maturity date on which the principle is repaid. In the absence of a
default by the issuer, bondholders can therefore wait for their investment to mature to get the
principle amount invested back. Bonds are thus more likely to be “buy and hold” assets, in
particular in the case of retail investors. In contrast, an equity investor who wants to retrieve
his investment has to rely solely on the secondary market, which makes the issues of investor
protection from abusive trading practices and price transparency more critical for equity
investors than for bond investors.

For the same structural reason, primary market purchases on the one hand and redemption
payments on the other account for the significant part of the buying and selling in the bond
markets, whereas secondary market trading of bonds tends to take place around the first
months or even weeks after issuance only. Buying and selling of shares, however, takes place
almost exclusively in the secondary markets.

12
  Asset swaps and switch trades present a problem for trade reporting. In essence, the asset swap spread or the
switch spread are the relevant economic criteria here. An issuer who buys back a structured or emerging
currency bond on an asset swap basis against simultaneous issuance of another but differently structured or
emerging currency bond on a liability swap basis presents even more complex questions of reporting.



                                                                                                              16
As we have already mentioned, bond prices can be linked through reference to government
bond and swap yields as well as through credit default swaps. This allows pricing engines to
price a whole range of bonds with limited benchmark input. This possibility does not exist to
the same extent with the equity market given the weaker correlations between individual
equities. In addition to this, equity markets are much more volatile than bond markets.

Finally, bond markets have since the 1980’s developed a bank dealer model -banks being
accustomed to lending are able and willing to take on inventories of bonds, particularly, if
they can hedge the interest rate risk. Interest rate hedging techniques developed in the 1980’s.
Equity markets have traditionally been broker markets with relatively lowly capitalised
brokers matching buyers and sellers. This also explains the levels of transparency traditional
in the two markets. The nature of equity markets means that brokers have relatively little
capacity to take on inventory and therefore need to find a matching counterparty for the trade
resulting in a greater need to advertise prices. Since equity brokers act on an agency basis,
they are less concerned about the negative impact of advertising prices. Fixed income dealers
have a greater ability to hedge and warehouse trades. That this ability to hold positions
enhances liquidity is undoubted. As we mentioned above, it was the development of hedging
markets, notably IRS but also the repo market, in the 1980’s that led to the development of
the bank dealer system at that time. That this enormously increased the capacity and liquidity
of the bond markets and reduced the cost of capital for borrowers is undoubted.

These structural differences and different models of transparency and liquidity have tended to
dictate that equities trade best on regulated exchanges with high degrees of transparency,
while bonds have tended to trade over the counter through dealers. There are of course areas
in which the two markets are converging. Large equity block trades are increasingly placed
with dealers via OTC trades whilst there is growing demand from smaller fixed income
investors for the kind of transparency provided by regulated exchanges. That the markets are
adapting in these ways argues against regulatory intervention.


7. Availability of and Payment for Post-Trade Data

Any potential problem with transparency may be not so much the availability of information
but its transmission to and collation by end users, in particular retail and smaller institutional
investors.

One of the main sources of income for regulated exchanges is the sale of data. A key issue in
a pre- or post-trade environment is how information is disseminated. There is some evidence
that given the current level of pre-trade transparency retail intermediaries are simply not
making the information available to their clients. Similarly, a mandatory post-trade
environment is only fully effective if data can be easily accessed at reasonable expense.
Many investors may only need access to trade data on a limited range of bonds either to carry
out research or as a price feed if they hold the bond in their portfolio. An investor who wants
to check whether he has obtained best execution may wish to obtain the price feed from


                                                                                               17
specific trading venues which his broker utilises for the period over which the trade was
executed.

We would recommend that providers of post-trade data be encouraged to make it available, on
line, in as digestible form as possible allowing buyers of data to select individual bonds or
groups of bonds on a tapered price basis. Easy forms of payment such as credit cards should
also be allowed. Should providers of post-trade data not be able to provide this service they
should be willing to sub-contact this provision to others including dealers and intermediaries.

Historical data (over three months old) should be made available to academic and other
research organisations at a reduced price.


8. Repo Markets and Liquidity Portfolios of Banks and Insurance
   Companies

The repo or secured lending market is growing rapidly in Europe and depends on its integrity
on reliable bond market pricing. Those who lend against the security of bonds will demand a
haircut depending amongst other things on the liquidity and price transparency of the bonds.
Central banks lead the market and determine the eligibility of bonds and other securities based
on a number of factors, usually credit ratings. The Bank of England in its eligibility criteria
for international securities requires that there be an ICMA price feed. Bank regulators will
usually base eligibility of securities for bank liquidity following central bank repo eligibility
criteria. Insurance company regulators determine solvency ratios based in part on the
liquidity of securities held to meet liabilities.

At the present time it is difficult to provide objective measures of the liquidity of a bond.
Greater availability of post-trade data should make this possible and regulators should be
encouraged to take these measures into account when setting haircuts and eligibility criteria.

Central bank repo eligibility and bank liquidity stock eligibility does much to improve the
liquidity of the bond market. It is not unreasonable that they should demand some degree of
post-trade transparency in return for these privileges. We note, however, that central banks of
the Eurosystem do not appear to demand pre-trade transparency as a price for eligibility.

We have mentioned above that the bond markets bank dealer model was developed in the
1980’s as a result of the introduction of interest rate hedging techniques. Of these the interest
rate swap market was key although government dealers had been permitted to take short
positions in stock subject to their capital constraints. The development of the swap market
depended firstly on the development of the term inter-bank deposit rate setting but the
development of a traded market depended on the ability of swap dealers to take short
positions in government bonds in order to hedge their positions (the so called swap
warehouse). This depended in the existence of a liquid repo market. It is remarkable that the



                                                                                              18
repo market, perhaps the most arcane and opaque of all financial markets, provides the key to
liquidity in all markets.


9. Macro (or Systemic) Transparency vs. Micro Transparency

AMTE has been privileged to have, as invitees on its working group looking at these issues,
two central banks; the ECB and the Banque de France. The ECB’s focus on systemic
liquidity and relevant transparency has brought a special perspective to our deliberations. It is
at least arguable that regulatory intervention should focus on the market superstructure rather
than on the microstructure where the consequences of regulatory intervention are less easy to
predict. Increased information on the more liquid CDS markets through published fixings
would contribute significantly.
The ECB in its paper “Implications for liquidity from innovation and transparency in the
European Corporate bond market” (August 2006) argues that published measures of net
exposures or concentrations of positions in the bond market would provide the most important
degree of transparency. They conclude their paper as follows:

 “We do not see how pre- or post-trade transparency can improve systemic liquidity. But
other forms of transparency may. The most promising avenue could be an enhanced
availability of data on net exposures or concentration of positions. This type of information
could indeed help market participants and competent authorities to properly value, manage
and price the increasing risks of homogeneous behaviour, which possibly causes crowded
trades, makes the market more vulnerable to shocks and thereby threatens systemic liquidity.”

AMTE endorses these findings and would encourage regulators to research further with
market participants how these recommendations could be put into practice.

It is worth recalling why and how market makers need to provide capital to a market making
activity. In essence, market makers are option writers, as was said above. In return for giving
others the option to hit their bids or lift their offers, they hope to capture the bid/offer spread
and have access to a privileged flow of information about market activity. Their risk is that
short term price variations increase and that, consequently, tight bid offer spreads dictated by
market competition no longer compensate for volatility and also, more importantly, that large
shifts in market prices leave market makers with significant losses on inventory 13 . It is to
protect against such large losses on inventory resulting from significant market shifts that
much dealer capital is or should be tied up. If dealers are further exposed by publication in
real time of their market positions then the cost of market making will be further increased. In
this context, please see Appendix 3 with some empirical evidence on the daily returns
distributions in the European bond markets.

13
  ECB Occasional Paper “Implications for liquidity from innovation and transparency in the European
Corporate bond market” discusses the evidence for and consequences of liquidity “black holes”. We will discuss
this paper further.



                                                                                                           19
An important rationale for regulatory intervention is to improve market/systemic stability and
thereby to improve the efficiency of capital that market makers are required to commit to
market making. This should not be by an artificial reduction in regulatory capital but by a
reduction in the real economic capital that market makers have to commit and a consequent
reduction in regulatory capital. A reduction in capital utilised by market makers would result
from successful efforts to improve systemic liquidity and this could be achieved by
improvements in what we refer to as systemic or macro transparency.

All investors, but particularly unsophisticated retail investors, are also likely to be better
served by measures that warn them of large positions being built up in the market than by
measures which improve marginally the price they obtain for individual transactions, i.e. by
increased macro rather than micro transparency. Individual market makers are unable to judge
to what extent individual market positions have been built up as some market participants
may trade with a large number of market makers. No one knows the overall positions built up
by individual players or by the market as a whole in the OTC market. Regulators should be in
a position to consolidate market maker positions and thus to estimate aggregate market
positions. The results of such studies should therefore be published and allow participants to
judge to what extent prices have been influenced by large positions. Such type of
transparency, which is supposed to improve liquidity under stress conditions and that we call
systemic or macro transparency as distinct from what we call micro transparency, is referred
to in the ECB Occasional paper. There is a good precedent for this in the data published by
regulated futures exchanges which show the aggregate open positions held.

Evidence of increased occurrences of “liquidity black holes” in financial markets is claimed,
that is instances of suddenly disappearing liquidity due to sharp price jumps resulting from all
market participants seeking to buy or sell at the same time. We believe that to the extent such
shifts are becoming more evident they are caused by more rapid and general dissemination of
market moving information (that is increased transparency of market information, both
exogenous and endogenous) and are easier to measure because of tighter bid/offer spreads. In
a perfect market where all information is immediately known by all market players any new
information should cause a stochastic jump in prices. Such a jump will be the more evident
the tighter the bid/offer spread in the market and the greater the price transparency of the
market. Compare this to a market with wide bid/offer spreads, low price transparency and
slow dissemination of information among market players. Some players will be continuing to
buy in the market while informed players are selling as a result of the new information. If the
bid/offer spread is wide the shift in prices may be no wider than the bid/offer 14 . In addition, if
prices are not published, it will take a while to detect that prices have moved.

The ECB paper argues, inter alia, that the measures to reduce the homogeneity of investor
behaviour in the bond markets will improve liquidity. AMTE agree with this conclusion.
Wider geographic and investor sector participation is good for bond markets. Increased retail
14
   In the late 1980’s a bid offer yield spread of 10 basis points in the 2/3 year interest rate swap market for large
size was common. In emerging currency bond and swap markets today bid/offer yield spreads can be more than
50 basis points.



                                                                                                                  20
participation resulting from increased macro and micro transparency will improve liquidity as
it has in the past.


10. Primary Bond Markets Transparency

When considering transparency in the primary bond markets, one should look at the different
phases of the new issue process and analyse the transparency required for the market and its
impact, particularly on investor protection. We define the primary or new issue bond market
to be the market for to-be-issued bonds from the first day when marketing activities for the
sale or distribution of the bonds start to the first settlement day of the bonds, i.e. the day on
which the bonds are effectively issued by the issuer, delivered to and paid for by the investors.
It should be noted that the primary bond market in Europe is almost exclusively an
institutional investor market, and private investors, in general, do not participate in primary
debt new issues.

During the first phase of the new issue process, the pre-pricing, or marketing, phase, lead
managers must ensure that any information given to the market concerning an offering is
accurate and timely delivered to potential investors. This information, prepared by the issuer
or the banks lead-managing the offer, is typically set out either in a stand-alone prospectus or
a final terms document prepared in connection with bonds issued off an offering programme –
typically a medium-term notes programme – for more frequent issuers and which is to be read
together with a base prospectus previously drawn-up by the issuer for such offering
programme. The mandatory informational requirements with regard to such a prospectus – be
it a stand-alone or a base prospectus – and relevant rules with respect to advertising the new
issue are laid down in the Prospectus Directive and in Regulation (EC) 809/2004 to the
Prospectus Directive. All such offering documents may only be distributed in accordance
with the relevant laws and regulations of the jurisdictions in which the offer of the bonds
takes place, e.g. in the case of the UK, in compliance with the Financial Services and Markets
Act. Investment firms involved in the primary sale or distribution of the bonds will typically
be bound to these specific country selling restrictions by way of a contractual agreement with
the issuer and these country selling restrictions will usually be included in the prospectus.
Lead managers must provide a final prospectus or a copy of the final terms together with the
base prospectus to all investors that buy, place an order or to whom the new issue was
marketed prior to the start of the marketing activities.

The next phase in the new issue process is bookbuilding, i.e. the phase from the day on which
the orderbook is opened for subscription of the bonds by investors to the date of pricing the
new issue (prior to which the book will have been closed). We note that the Prospectus
Directive requires the issuer or the offeror(s) to include in the prospectus or the final terms
document the final offer price for the bonds. If the new issue has not been priced yet, then the
conditions in accordance with which the final offer price will be determined or a maximum
price has to be specified in these documents. Otherwise, investors are entitled to withdraw



                                                                                              21
their orders for subscription. The final offer price has then to be disclosed to the public and
filed with the competent authority.

In addition, Directive 2003/6/EC on insider dealing and market manipulation (market abuse)
(together with the relevant implementing directives and regulations, the “Market Abuse
Directive”) controls the information given to the market by the lead managers during the
bookbuilding phase. The lead managers cannot provide misleading or inaccurate information
about any aspect of the new issue, including the size of the order book, the price sensitivity of
the order book or anything else that may give a false impression about the success or
otherwise of any new bond offering. This also applies to comments made by representatives
of lead managers in the press before and after pricing of the issue.

The Market Abuse Directive also regulates the transparency post-pricing of a bond new issue.
In particular, it defines safe harbour rules regarding the ability of lead managers to stabilise,
or provide price support for, a new issue which require that if any stabilisation is undertaken
by the lead managers, then it must be documented, disclosed to the public and kept on record,
and a notification must be sent to the relevant competent authority. In particular, the price
range within which stabilisation was carried out for each of the dates during which stabiliation
transactions have been undertaken has to be adequately disclosed to the public.

Given the extensive regulatory requirements of both the Prospectus Directive and the Market
Abuse Directive, we think that transparency issues in the European primary bond markets are
sufficiently covered by the existing legislation.




                                                                                              22
Appendix 1: Key Points Arising from AMTE Response to FSA
 Discussion Paper 05/05 on Trading Transparency in the UK
                  Secondary Bond Market




                                                            23
Key Points Arising from AMTE Response to FSA Discussion Paper 05/05 on Trading
Transparency in the UK Secondary Bond Market


   •   AMTE strongly support a pan-European vision

   •   Wholesale and retail markets present different models of transparency. No market
       failure is observed in the wholesale market as a result of a lack of transparency,
       though the retail market presents scope for improvement.

   •   Consolidation of currently available data would improve the situation for smaller
       institutional investors and for retail. An important issue is who provides data, how it
       will be disseminated and whether on a free (or at least low cost) basis.

   •   Bonds differ from equities in that bonds can be priced on government bonds, IRS and
       increasingly CDS.

   •   Liquid bonds are more tightly priced and transparent than illiquid bonds (please note
       the causality).

   •   Best execution is more easily obtained in liquid markets and AMTE question whether
       full transparency or public dissemination of post-trade transparency is necessary to
       gauge best execution.

   •   The fact that portfolio valuation can be better carried out if more price transparency is
       available, does not provide evidence of market-failure and does not call for MiFID
       style equity pre- and post-trade transparency requirements. (N.B. the FSA asked for
       evidence of market failure; the term is not mentioned in the EC call for evidence).

   •   Retail represents a small section of the European bond market and retail generally own
       bonds on a buy and hold basis. Secondary markets provide an exit for retail.
       Suitability and best execution provisions are the best way to protect retail.
       Appropriate price transparency should be targeted to ensure best execution. Parmalat
       and Argentina were nothing to do with price transparency. Given the small size of the
       retail market regulation should be targeted and measured.

   •   Retail could rely on the burden of proof that the implementing measures of Art. 21(5)
       of MiFID will impose on investment firms.

   •   Tighter spreads are observed in the Euro market than in the US, even after
       implementation of TRACE. The greater number of MTFs in Europe results in greater
       availability of pre- and post-trade information.




                                                                                             24
•   Conclusion:

       - There is no evidence of market failure in the wholesale bond market.

       - Cost benefit analysis should, in addition to focusing on a reduction of
         commitments of market makers to illiquid markets, consider whether liquidity
         will be driven away from cash markets into the more opaque markets such as
         CDS.

       - The proliferation of data is a problem. ICMA TRAX reporting system could be
         an interesting avenue to redress this problem.




                                                                                  25
Appendix 2: AMTE Response to FSA Discussion Paper 05/05 on
   Trading Transparency in the UK Secondary Bond Market




                                                             26
        AMTE RESPONSE TO FSA DISCUSSION PAPER 05/5 ON TRADING
          TRANSPARENCY IN THE UK SECONDARY BOND MARKET
                          THEMATIC PAPER

                                                                                  January 2006

AMTE

AMTE, the Euro Debt Market Association, was created in July 2002 and counts now nearly
60 members, leaders in their field and representative of the euro debt markets: issuers,
financial intermediaries, investors and other market professionals.

The association aims to facilitate the increase of the depth, liquidity, transparency and
innovative nature of the euro debt markets. In this context, AMTE’s purpose is to contribute
to the development and effective operation of euro fixed income and derivatives markets by
supporting the various professionals in these markets in their efforts to come to an agreement
on concrete ways to reach these goals.

To achieve its objectives, AMTE conducts work and research, and organises consultations
among its members on all subjects relating to any type of operation on euro-denominated
financial markets excluding equity markets.

In November 2005, AMTE established a Working Group dedicated to bond market
transparency in Europe with a view to providing input in the context of the Markets in
Financial Instruments Directive (MiFID) Art. 65 review. This Working Group comprises
representatives of the three constituencies of the market (issuers, banks and investors). The
scope of the Working Group will focus on vanilla bonds at the first stage. This Working
Group will likely be in existence for some time as the MiFID for bonds process will take a
number of years before implementation. The Working Group will remain available to
respond as AMTE sees fit to the ongoing debate on MiFID for bonds.

The following thematic paper prepared by the AMTE Working Group has been agreed by the
AMTE Board and endorsed by the institutions, members of AMTE (see attached lists of
AMTE Working Group members, AMTE Board members and AMTE institution members).

Overview

We welcome the FSA’s initiative to engage market participants at an early stage in order to
gather information and opinions in view of the upcoming MiFID Art. 65 review.

We believe that the FSA is approaching the issue from the right angle in asking the threshold
questions “Are there any market failures in the bond markets? If so, … how do they arise?”.
As the FSA has also recognised, correctly in our view, a regulatory initiative should not be the
first response even if a market failure was identified. Other avenues such as self-regulatory



                                                                                             27
industry led initiatives should be explored and only if the answer to the latter question
identified a shortfall in the amount or quality of transparency as the cause of any such failures
should such intervention be directed at improving transparency. And if it was determined that
regulatory intervention was the best or only solution, such regulatory intervention should be
subject to a rigorous cost–benefit analysis.

We would like to suggest that the FSA should not look for one single answer to these
questions, as there is no one single bond market in Europe. Different markets are demarcated
by features such as the types of products, the level of liquidity, the prevailing trading media,
different currencies and the types of participants. These markets extend beyond national
boundaries and, whilst some features could be prevalent in certain jurisdictions (as, for
example, is the case of direct retail holdings of bonds in Italy), there are no structural
differences that would justify geographic fragmentation. For this reason, we strongly support
a pan-European vision of the bond market and do not think that the debate on transparency in
the secondary bond markets should produce different outcomes in the various European
jurisdictions. However, we recognise that the European wholesale and retail markets in
particular present distinctive features and substantially different models of transparency. As
will be outlined below, the AMTE Working Group concluded that no market failure can be
observed in the wholesale bond markets that could be attributed to a lack of, or insufficient,
transparency. The retail market presents some scope for improvement, but it should be
further investigated whether such improvements should solely, or mainly, be sought in the
area of secondary markets’ price transparency.

Indeed, numerous resources are available to bond markets’ participants providing varying
degrees of pre-trade and post-trade transparency 15 . Dealers’, brokers’ and institutional
investors’ representatives in the AMTE Working Group confirmed that the current level of
pre and post-trade transparency meets their requirements.

However, we recognise that the lack of consolidation of available market information, such
that different data sets are provided by different information services, in most cases at a cost,
can result in access to a broad range of data being costly for smaller institutional investors and
prohibitive for retail. We believe that further consolidation of currently available trading
information would mark a meaningful improvement in secondary bond market transparency
and may facilitate wider access to trading data. However an important issue will need to be
addressed as to who will provide such information, how it will be disseminated and whether it
is on a free (or at least low cost) basis.

We broadly agree with the FSA’s analysis that externalities and information asymmetries
resulting in distorted price formation and failure of best execution would indicate a market

15
  Single dealer and multi-dealer (to customer) electronic trading platforms provide a combination of executable
and indicative quotes and real-time or end of day post-trade prices. Data vendors disseminate dealer quotes,
exchanges’ and electronic trading platforms’ post-trade information. A detailed picture of the main price
information sources currently available is offered by The Bond Market Association, European Bond Pricing
Sources and Services: Implications for Price Transparency in the European Bond Market, (2005).



                                                                                                            28
failure. However, we concur with the FSA’s preliminary findings that there is no evidence of
such indicators of market failure.

As regards price formation, it should be noted that the importance of price data in the bond
markets is not comparable to its importance in the equity market. Bonds can be priced more
easily than equities and based on observable factors (benchmark government bonds, the
interest rate swap curve and, increasingly, credit derivatives). Trading information is
therefore less central to the pricing of bonds.

That said, the current price formation process for bonds appears to be efficient and in line
with market dynamics. The liquid segments of the bond markets are tightly priced and more
transparent due to a number of factors: the issues are of larger sizes, they are more suitable for
(cheaper) electronic trading, they are more commoditised, there is more competition and
committing capital is less costly. Conversely less liquid securities are more expensive to
trade as they have smaller issue sizes and fewer dealers are willing to trade and commit their
capital. This also results in illiquid markets being less transparent. Similar considerations
apply to pre-trade information. In liquid segments of the market it is easier to find firm
quotes, in illiquid segments of the markets it is more common for quotes to be only indicative.
In the latter, dealers are prepared to service their client base by making a market in the bonds,
but they are reluctant to expose their quotes to the market at large as this may result in their
competitors exploiting such liquidity.

As competition is arguably the main driver of efficient price formation, it is no surprise that,
as found by the FSA in discussions with market participants, best execution is more easily
achieved in more liquid markets. It is also in line with the considerations made in the
preceding paragraph that less liquid, less transparent markets may reduce the ability to
monitor best execution. We question, however, whether full transparency or public
dissemination of post trade data are necessary in order to gauge best execution. We share the
FSA’s view that optimal transparency need not be maximum transparency.

There is no question that portfolio valuations could be executed more accurately and
efficiently if price information was readily accessible to institutional investors. However, this
does not provide evidence of market failure and does not call for a transparency model such
as that mandated by MiFID for shares.

So that the need for regulatory intervention can be accurately assessed, it is important that the
debate regarding transparency in the secondary bond markets remains focussed on the
threshold questions referred to above and that transparency is not treated as a panacea to
administer generously on the assumption that it could not cause harm.

Retail vs. Wholesale

We invite great focus on the fundamentally different features of the retail and wholesale
secondary bond markets. As observed by the FSA, direct holdings of bonds by retail



                                                                                               29
investors (other than high net-worth individuals) represent a small section of the European
bond market. Importantly, direct holdings do not translate into direct trading by retail
investors in the market whether at the entry or exit point. Retail investors ordinarily hold
their investments to maturity, the secondary bond market provides retail investors with an exit
route should they choose to divest but by and large retail investors do not otherwise engage in
active trading in the market 16 .

Currently, retail access to the secondary bond markets is mainly indirect and the retail activity
in bonds is predominantly buy-and-hold. It is important to note this in the determination of
potential improvements to transparency in the retail secondary bond market. Stringent
suitability requirements and effective implementation of the best execution obligations
imposed on retail brokers should play a central role in the protection of retail investors.
Improved dissemination to retail investors of price information could empower them to
monitor the quality of execution received from their brokers. The appropriate amount, quality
and sources of price transparency should be identified by targeted measures in order to fulfil
best execution monitoring objectives 17 .

The debate on transparency in the retail markets has at times become intertwined with
concerns over recent issuers’ defaults, such as that of Parmalat and the Argentinean
government. However, there is no evidence that any amount of price transparency could have
prevented, or could even just have alerted the public to, such events 18 .

Given the more limited price information resources that retail investors have access to, and
even though this is not in itself indicative of a market failure, there could be a stronger case
for targeted regulatory intervention to improve transparency in the retail market. Yet, if the
ultimate objective is the protection of retail investors, then education, upfront disclosure 19 ,
suitability, best execution and principal protection should play important roles if not leading
ones.



16
   Cf. TRACE trading data, analysed, inter alia, by Jean-Pierre Casey and Karel Lannoo, Europe’s Hidden
Capital Markets: Evolution, Architecture and Regulation of the European Bond Market, CEPS (2005), where it
is observed that retail trades represent 1.8% in value terms and that the situation of the European market mirrors
such data. The authors further observe that these data rely on the determination that a retail trade is any trade of
a nominal value of $ 100,000 or less and question the accuracy of this definition inviting caution in the reading
of these figures.
17
   A sovereign issuer noticed that it is not always possible for the retail investors to secure the liquidity of their
bond investments; traditional “stock-exchange type” regulated markets, if coupled with market making
arrangements, seem to provide an adequate base for the provision of retail market liquidity, within the existing
regulatory framework, and should be supported by the community of issuers, intermediaries and regulators.
18
   Casey and Lannoo identify a key element of retail investors’ protection in principal protection. They correctly
point out that retail investors view fixed income investments as ones which provide a steady cash flow without
risk of losing their capital and that principal protection is eminently a corporate governance and accounting
issue.
19
   Within the AMTE Transparency Working Group it has been suggested that the issuer or lead manager should
state at the time of listing whether the bond will be subject to post-trade transparency.



                                                                                                                  30
In addition to qualitative differences, the relative size of the retail and wholesale secondary
bond markets should also present a strong case for targeted and measured regulatory
intervention (if any should be deemed necessary).

Best execution

As regards best execution, we believe that the debate should focus on the post-MiFID
implementation environment, an important component of which is its Art. 21(5). In such an
environment, investors’ protection would rest essentially on upfront disclosure of, and
investor consent to, the firm’s execution policy and on the firm’s obligation to “be able to
demonstrate to their clients, at their request, that they have executed their orders in
accordance with the firm’s execution policy”. As further specified in the proposed MiFID
implementation measures (albeit still in draft form), “when an investment firm executes an
order on behalf of a retail client, the most important factor for determining the best possible
result as required by Art. 21(1) shall be the total consideration (representing the price of the
financial instrument and the costs related to execution) payable by the client” 20 . Whilst the
investment firm will have access to the information disseminated in the wholesale market and
therefore to a sufficiently representative spectrum of trading opportunities and will in turn
have obtained best execution from its brokers, the investor could rely on the burden of proof
that the implementing measures of Art. 21(5) of MiFID will impose on the investment firm.

Arguably best execution should not bear the entire burden of ensuring that retail investors are
able to deal on fair terms, but if effectively implemented, it could go a long way towards
achieving this objective.

Lessons from the U.S.

As much of the current debate on secondary bond markets transparency is informed by the
observation of the U.S. experience, we would like to point out a fundamental difference
between the U.S. and the EU markets which should suggest caution in introducing U.S. style
regulatory requirements in the European markets. Market failures as might have been
perceived in the U.S. secondary bond markets could be explained by looking at the different
competitive environments. In Europe over twice as many dealers as in the U.S. compete to
attract business in a much smaller market. Such intense competition is a strong inhibitor of
market failures. Tighter spreads have been observed 21 in the Euro market than in the U.S.,
even after the implementation of TRACE 22 .



20
   Art. 18(2) Working document ESC/23/2005/Rev 2 – Draft Commission Working Document on conduct of
business rules, best execution, client order handling rules, eligible counterparties, clarification of the definition
of investment advice and financial instruments.
21
   Casey and Lannoo.
22
   Trade Reporting And Compliance Engine launched in July 2002 by the National Association of Securities
Dealers (NASD).



                                                                                                                 31
In addition, a much higher number of multi-lateral trading platforms in Europe results in
greater availability of pre and post-trade information.

Concluding remarks

As the above remarks will have shown, AMTE agrees with the FSA’s preliminary finding that
no market failure can be observed in the wholesale secondary bond markets that calls for
regulatory intervention.

We also agree that a rigorous cost-benefit analysis should be conducted before mandating
specific pre and post-trade transparency. The nature of the costs that could be associated with
regulatory intervention is correctly identified by the FSA in explicit and implicit costs.
However, in particular as regards the implicit costs, the FSA appears to focus on the potential
reduction in the commitments of market makers to less liquid markets. Large investors have
confirmed to the AMTE’s Working Group that they also would have concerns with greater
transparency especially when trading in large size 23 . It is a concern of the AMTE that
liquidity could be driven away from the cash market into more opaque and less transparent
sectors of the market such as the credit derivatives market. The AMTE Working Group fears
that this could have the effect of exacerbating the existing imbalance between the respective
sizes of the cash and of the credit derivatives markets.

Finally the AMTE Working Group recognises that the proliferation of data available in the
market from a variety of sources can be a barrier to some constituencies’ access to price
information. ICMA TRAX reporting system could be an interesting avenue to address this
situation. We conclude that further consolidation of currently available pre and post-trade
information could address such concerns. The AMTE Working Group also concludes that
any changes to the architecture of the secondary bond markets must be carefully researched
such that any transparency measures designed to assist the retail market do not disrupt the
smooth functioning of the wholesale market. If any measures are considered, the AMTE
Working Group urges that the following are taken into account:

     •   there should be consultation with the industry;
     •   a cost-benefit analysis should be conducted;
     •   regulation should be targeted and measured to avoid being counter productive;
     •   pan-European regulatory requirements should be preferred to national ones;
     •   any regulatory requirements should be phased in;


23
   Casey and Lannoo observe that “increasing transparency increases the costs of liquidity provision and the
costs for investors to unwind large positions. For example, the introduction of the TRACE post-trade reporting
system in the US has lowered transaction costs for retail investors making direct investments. However, it may
have increased transaction costs for institutional investors. Since the vast majority of retail investments in fixed
income are channelled through funds, such policy measures designed by regulators to protect retail investors
might paradoxically damage the very interests they are designed to protect, as funds pass on higher transaction
costs to their retail clients, e.g. in the form of lower returns.”



                                                                                                                32
•   the U.S. market experience (TRACE) should be carefully, and independently,
    analysed. The European market is different from the U.S. and caution should be
    exercised before regulation is imported.




                                                                               33
AMTE Working Group « Bond market transparency in Europe » members


  •   Ziad Awad, Goldman Sachs
  •   Aurélie Bedouet, AMTE
  •   Charles Berman, Citigroup
  •   Valérie Blanchin, AMTE
  •   Eric Brard, Société Générale Asset Management
  •   David Clark, EIB
  •   Alban de Clermont-Tonnerre, Goldman Sachs
  •   Stefan Ericsson, ABN Amro
  •   James Garvey, Goldman Sachs, chair of the AMTE Working Group
  •   Françoise Guillaume, Société Générale Asset Management
  •   Kenneth G. Lay, The World Bank
  •   Matthieu Louanges, Pimco
  •   Lisa Rabbe, Goldman Sachs
  •   Rodolphe Sahel, Société Générale
  •   Ivan Zelenko, The World Bank




                                                                     34
                               AMTE Board members

Chairman:
René Karsenti
Director General of Finance of EIB (European Investment Bank)

Treasurer:
Jean-Pierre Mustier:
CEO in charge of Corporate and Investment Banking activities of Société Générale

Secretary:
Jean-François Boulier
Head of Euro Fixed Income and Credits at CAAM (Crédit Agricole Asset Management)

Board members:
Charles Berman
Managing Director, Co-Head of Fixed Income Capital Markets, Europe at Citigroup

Joe Dryer
Managing Director, Global Co-Head of Corporate Finance and Origination at DrKW
Germany of Dresdner Bank AG

Heinz W. Fesser
Head of Fixed Income Germany of DWS/DeAM and Managing Director of DWS Investments

James Garvey
Head of Debt Capital Markets, Fixed Income & Forex Markets Division at Goldman Sachs

Robert B. Gray
Chairman, Debt Finance & Advisory of HSBC Bank plc and Vice-Chairman of ICMA
(International Capital Market Association)

Matthieu Louanges
Head of the European Government Team of PIMCO, generalist Portfolio Manager

Bertrand de Mazières
Chief Executive of Agence France Trésor

Gerhard Schleif
Managing Director of Bundesrepublik Deutschland Finanzagentur GmbH

Secretary General:
Valérie Blanchin




                                                                                       35
                          AMTE institution members


• ABN Amro                              • ICAP Electronic Broking
• AFT (Agence France Trésor)            • IXIS Corporate & Investment Bank
• AFTI                                  • JP Morgan
• Axa Investment Managers               • KfW
• Barclays Bank                         • LCH.Clearnet
• Banque de France                      • Lehman Brothers
• Bloomberg                             • MarketAxess
• BNP Paribas                           • Merrill Lynch
• BNP Paribas Asset Management          • Moody’s
• Bundesrepublik Deutschland            • Morgan Stanley
Finanzagentur                           • MTS Group
• CAAM (Crédit Agricole Asset           • Natexis Banques Populaires
Management)                             • Nomura
• CADES (Social Security Debt           • PIMCO
Repayment fund)                         • S.G. Asset Management
• Calyon                                • Société Générale
• CdF                                   • The Royal Bank of Scotland
• CIF Euromortgage                      • The World Bank
• Citigroup                             • TradeWeb
• CNO (Bond Standardisation             • UBM
Committee)                              • UBS Warburg
• CME (Chicago Mercantile Exchange)     • Véolia Environnement
• Crédit Foncier de France              • Viel & Cie
• Crédit Suisse First Boston
• DEPFA Bank Plc
• Deutsche Bank
• Dexia Crédit Local
• Dresdner Bank
• DWS Investments
• EIB (European Investment Bank)
• ETC Pollak Prebon
• Eurex
• Euroclear
• Eurohypo AG
• Euronext
• Goldman Sachs
• Hellenic Republic – Public Debt
Management Agency
• HSBC
• HVB


                                                                             36
Appendix 3: Empirical Evidence on the Daily Returns
    Distributions in the European Bond Markets




                                                      37
Empirical Evidence on the Daily Returns Distributions in the European Bond
Markets

When analysing the relationship between market makers’ willingness to commit
capital for trading and the expected return on capital committed (bid/offer spreads), it
is worth considering the nature of most financial markets. Far from being normal, the
distribution of returns will usually display many more observations at the middle of
the distribution than implied by a normal distribution curve, but will also show a
greater frequency of extreme price fluctuations (“fat tails”) than implied by a normal
curve distribution. Such price distribution curves are called “leptokurtic”. Figure 1,
which shows the distribution of daily returns for the Dow Jones index for the period
1915-2001, illustrates and compares a leptokurtic curve with a curve that is normally
distributed.

Figure 1




                                  Source: Bloomberg

    Mean 0.0002995705
  StdDev 0.011369106
    Skew -0.269016617
 Kurtosis 20.70573405


At the centre of the distribution, the tendency for very short term price variations to
be smaller than implied by a normal distribution could encourage tighter bid/offer
                                                                                    38
spreads than would normally be justified. It may even be caused by market maker
behaviour.

Similar patterns but lower kurtosis are displayed using long time series data provided
by the Federal Reserve Bank of St. Louis for constant maturity 5-year and 10-year US
treasuries for the period 1962-2006.


Figure 2
                     5-year UST CMT Daily Price Returns Distribution
                                     Jan 1962 - Aug 2006
               25


               20


               15
           %




               10


               5


               0
                    0%

                      %

                      %

                      %

                      %

                      %

                      %

                      %




                     %
                     %

                     %
                     %

                     %

                     %

                     %
                     %

                     %




                   50

                   00

                   50

                   00
                   50

                   00

                   50

                   00
                   00
                  .50

                  .00

                  .50

                  .00

                  .50

                  .00

                  .50
                  .0




                3.

                4.
                2.

                2.

                3.
                0.

                0.

                1.

                1.
               -4

               -3

               -3

               -2

               -2

               -1

               -1

               -0




                           Source: Federal Reserve Bank of St. Louis

 Mean               3.83E-05
 Standard
 Deviation          0.010732
 Skew                0.00374
 Kurtosis           7.161525




                                                                                   39
Figure 3
                         10-year UST CMT Price Returns Distribution
                                      Jan 1962 - Aug 2006
                 30

                 25

                 20
             %




                 15

                 10

                 5

                 0
                      0%

                      0%

                      0%

                      0%

                      0%

                      0%

                      0%

                      0%

                       %

                       %

                       %

                       %

                       %

                       %

                       %

                       %

                       %
                     00

                     50

                     00

                     50

                     00

                     50

                     00

                     50

                     00
                    .0

                    .5

                    .0

                    .5

                    .0

                    .5

                    .0

                    .5
                  0.

                  0.

                  1.

                  1.

                  2.

                  2.

                  3.

                  3.

                  4.
                 -4

                 -3

                 -3

                 -2

                 -2

                 -1

                 -1

                 -0




                         Source: Federal Reserve Bank of St. Louis
 Mean                 2.27E-05
 Standard
 Deviation             0.00883
 Skew                 -0.00182
 Kurtosis             4.912295


Trying to replicate this using European bond market data provided by iBoxx for the
period from 1 January 2005 to 15 August 2006, presents a very different result.




                                                                               40
Figure 4
                       iBoxx Aggregate Index Daily Returns Distribution

                  30

                  25

                  20
             %




                  15

                  10

                  5

                  0
                                                        %
                      %


                              %


                                       %


                                                %




                                                               0%


                                                                      1%


                                                                             2%


                                                                                     3%


                                                                                            4%


                                                                                                   5%
                                                      .1
                    .5


                            .4


                                     .3


                                              .2




                                                             0.


                                                                    0.


                                                                           0.


                                                                                   0.


                                                                                          0.


                                                                                                 0.
                  -0


                          -0


                                   -0


                                            -0


                                                    -0




                                  Source: International Index Company

 Mean                      8.96E-05
 Standard
 Deviation                 0.001673
 Skew                      0.019305
 Kurtosis                  -0.13903

Figure 5

                       iBoxx EUR Index Daily Returns Distribution

             25

             20

             15
       %




             10

              5

              0
                                                                                                      5%
                 %

                           %

                                   %

                                                %

                                                         %

                                                               0%

                                                                      1%

                                                                              2%

                                                                                       3%

                                                                                              4%
               .5

                         .4

                                 .3

                                            .2

                                                     .1

                                                             0.

                                                                    0.

                                                                            0.

                                                                                     0.

                                                                                            0.

                                                                                                    0.
             -0

                       -0

                               -0

                                           -0

                                                    -0




                                      Source: International Index Company
                                                                                                           41
 Mean                   8E-05
 Standard
 Deviation              0.001888
 Skew                   0.068715
 Kurtosis               -0.07896


Figure 6

                     iBoxx USD Index Daily Returns Distribution

             25

             20

             15
       %




             10

              5

              0
                                                                                            5%
                 %

                         %

                                 %

                                          %

                                                   %

                                                         0%

                                                                1%

                                                                       2%

                                                                              3%

                                                                                     4%
               .5

                       .4

                               .3

                                      .2

                                               .1

                                                       0.

                                                              0.

                                                                     0.

                                                                            0.

                                                                                   0.

                                                                                          0.
             -0

                     -0

                             -0

                                     -0

                                              -0




                             Source: International Index Company

 Mean                   8E-05
 Standard
 Deviation              0.001888
 Skew                   0.068715
 Kurtosis               -0.07896




                                                                                                 42
Figure 7

                  iBoxx GBP Index Daily Returns Distribution

             18
             16
             14
             12
             10
        %




             8
             6
             4
             2
             0
                  %
                  %
                  %
                  %
                  %
                  %
                  %
                  %
                  %


                 1%
                 2%
                 3%
                 4%
                 5%
                 6%
                 7%
                 8%
                 9%
                 0%
                .9
                .8
                .7
                .6
                .5
                .4
                .3
                .2
                .1




              0.
              0.
              0.
              0.
              0.
              0.
              0.
              0.
              0.
              0.
             -0
             -0
             -0
             -0
             -0
             -0




             -0
             -0
             -0




                        Source: International Index Company

 Mean                 0.000177
 Standard
 Deviation            0.002572
 Skew                 0.110433
 Kurtosis             0.338919


If a longer period, 2000-2006 is used, however, a slightly different picture emerges.




                                                                                        43
Figure 8

                    iBoxx Aggregate Index Daily Total Returns Distribution
                                          2000 to 2006
            25


            20


            15
       %




            10


                5


                0


               8%
               1%
               2%
               3%
               4%
               5%
               6%
               7%


               9%
               0%
               0%
           -0 %
           -0 %
           -0 %
           -0 %
           -0 %
           -0 %
           -0 %
           -0 %
           -0 %
                %
              .0
              .9
              .8
              .7
              .6
              .5
              .4
              .3
              .2
              .1



            0.
            0.
            0.
            0.
            0.
            0.
            0.
            0.
            1.
            0.
            0.
           -1




       Source: International Index Company

Figure 9

                         iBoxx EUR Index Daily Total Returns Distribution
                                    2000 to 2006
           25


           20


           15
       %




           10


           5


           0
               0%
               1%
               2%
               3%
               4%
               5%
               6%
               7%
               8%
               9%
               0%
                %
                %
                %
              .7%


                %
                %
                %
                %
              .6%




                %
              .0
              .9
              .8




              .5
              .4
              .3
              .2
              .1
            0.
            0.
            0.
            0.
            0.
            0.
            0.
            0.
            0.
            0.
            1.
           -1
           -0
           -0
           -0
           -0
           -0
           -0
           -0
           -0
           -0




                             Source: International Index Company




                                                                             44
Figure 10

                         iBoxx USDIndex Daily Total Returns Distribution
                                    2000 to 2006
            25


            20


            15
       %




            10


            5


            0
               0%
               1%
               2%
               3%
               4%
               5%
               6%
               7%
               8%
               9%
               0%
            -0 %
            -0 %
            -0 %
            -0 %
            -0 %
            -0 %
            -0 %
            -0 %
            -0 %
              .1%
              .0
              .9
              .8
              .7
              .6
              .5
              .4
              .3
              .2

             0.
             0.
             0.
             0.
             0.
             0.
             0.
             0.
             0.
             0.
             1.
            -1




                             Source: International Index Company
Figure 11

                     iBoxx GBP Index Daily Total Returns Distribution
                                           2000 to 2006
             20
             18
             16
             14
             12
       %




             10
                 8
                 6
                 4
                 2
                 0
                 0%

              0. %
                 2%
                 3%
                 4%
                 5%
                 6%
                 7%
                 8%
                 9%
                 0%
             -0 %
             -0 %
             -0 %
             -0 %
             -0 %
             -0 %
             -0 %
             -0 %
             -0 %
              0. %

                 1
                .0
                .9
                .8
                .7
                .6
                .5
                .4
                .3
                .2
                .1

              0.


              0.
              0.
              0.
              0.
              0.
              0.
              0.
              1.
             -1




                             Source: International Index Company


                                                                           45
            iBoxx Index 2000-2006 Daily Total Returns
            Distribution


                               USD          EUR           GBP          Aggregate

            Mean               0.000256 0.000217 0.000247 0.0555%
            Standard           0.002618 0.0019   0.002841 0.002412
            Deviation
            Skew               -0.31802 -0.41492 -0.20185 0.208172
            Kurtosis           1.322117 1.049718 0.605718 1.11434
            Number of          1675     1675     1729     1729
            Observations
            Max                0.009935     0.005929      0.009612     0.9612%
            Min                -0.01122     -0.00808      -0.01057     -0.9777%
            Range              0.021159     0.01401       0.020179     1.9389%
            Range/StDev        8.082814     8.082814      7.103248     8.03795


Thus, short term European bond data 2005-2006 shows, if anything, negative kurtosis
(platokurtic).

The distributions for the period 2000-2006 are more normal than for the period 2005-
2006 and, with the exception of GBP, show a degree of kurtosis.

Longer term data is almost certain to show increased kurtosis as market events such
as LTCM come more fully into the data. This is clearly shown for very long term
Dow Jones data and for the less long term US Treasury data (1962-2006).
If this is the case, it strongly argues against use of short period volatilities for risk
management purposes. Leptokurtosis in markets is probably caused by the existence
of fat tail events 24 . Availability of long term market data will significantly improve
the ability of risk managers to assess risk of fat tail events.

One curious aspect of the iBoxx data is that GBP is platokurtic while USD is
leptokurtic and EUR data somewhere in between. The GBP market is well known for
being dominated by a relatively small group of highly professional investors. That
such a homogeneous and relatively small market may be showing less fat tail events
is interesting. The FSA feed back paper has already indicated wider bid/offer spreads
in the GBP market than in the EUR market and it is generally considered to be less
liquid and less transparent than either the EUR or USD markets. This perhaps merits
further research. Relative credit quality, average issue size and duration may also
have an impact.

24
TP   We are grateful to Chris Golden, Chairman of EFFAS-EBC for his contribution to this analysis.
                                                                                                     46
Appendix 4: Response to Specific Questions Raised by the
                European Commission




                                                           47
Question 1: Do you have any comment on the proposed scope of the Report?

Care should be exercised in consideration of differential transparency requirements
that liquidity is not driven from the market with the more onerous requirements to a
market with less onerous or no transparency requirements.

We also believe that more onerous reporting requirements placed on high quality and
liquid bonds may divert retail selling efforts to less onerous reporting regime in spite
of the suitability requirements imposed by MiFID.

Notwithstanding the above, we recognise that the nature and infinite number of
potential structures of OTC derivatives makes post-trade transparency in this sector
almost impossible in practice. In addition, many structures are proprietary.

We believe that many of the problems of transparency relate not to the absence of
information but to its availability to, and collation by, retail and smaller institutional
investors. Regulation, or perhaps self-regulation, of intermediaries designed to
improve data availability may have a greater beneficial impact than regulation of
dealers themselves.

Question 2: Do you consider this classification scheme to be sufficient for the
purposes of the review?

“Supranationals” should be taken together with “government guaranteed agencies”. In
order to avoid too many classifications these could be included together with
government bonds.

The Category “Credit Derivatives” should be renamed “Interest Rate and Credit
Derivatives”. Alternatively, the two categories should be separated.

Covered Bonds represent the largest sector of the European corporate bond market
and should have their own sub-section.

We emphasise that we recommend that differential regulatory environments for the
various cash markets should not be applied.

The categorisation of bond markets is by issuer. We wonder whether it is not
advisable to consider a separate classification by type of instrument; for instance, high
quality issuers including governments may issue smaller-sized structured and
emerging currency bonds which, by nature, will show a significantly lower degree of
liquidity and price transparency compared to large-sized plain-vanilla benchmark
bonds issued by the same issuer.


                                                                                       48
It is not clear whether “Cross Currency Swaps” are included under the category
“Foreign Exchange Derivatives” or under “Interest Rate Swaps”. Generally we
would prefer that “Cross Currency Swaps” and forward foreign exchange contracts,
with a maturity over 1-year, should be taken together with “Interest Rate Swaps”.

Question 3: Do you consider there are possible policy rationales for mandatory
transparency we have not listed?

With respect to “Market Efficiency”:

We would like to add the following sub-bullet point:
“mandatory transparency with regard to net positions by instrument and/or risk type
and to net positions by instrument and trading venue can mitigate reinforcing adverse
price movements in situations of distressed markets”

We question the assumption that increased transparency will lead to tighter bid/offer
spreads. Under certain conditions increased transparency may lead to wider bid/offer
spreads. This may or may not be a reasonable price to pay for increased transparency.
It is possible that post-MiFID best execution trading venues will display wider
bid/offer spreads than have hitherto been available.

With respect to “Response to technological developments”:

We believe that technological developments, taken together with the Best Execution
requirements of MiFID, have largely obviated the need for mandatory pre-trade
transparency. We do not therefore agree that this represents, in general, a policy
rationale in favour of mandatory transparency.

Question 4: Do you agree with our proposals for prioritisation of the review?

See our response to Question 1. We reiterate that differential regulation is likely to
cause a shift in liquidity to the less regulated categories. There are of course
considerable problems in mandating transparency for OTC derivatives where an
infinite number of structures may be created. We believe that, to the extent that pre-
trade transparency is deemed necessary to police best execution, the Commission
should work with the relevant trade associations to establish published benchmark
pricing preferably in the form of periodic rate settings taking an average of firm
dealer quotes (high and low quotes excluded). Post-trade transparency would create
problems of a major order. How would complex non-standard trades be reported?
The effort involved in establishing reporting mechanisms would have a seriously
debilitating impact on the market. Analysing the results would be possible only for
the most sophisticated players.


                                                                                   49
We share the view of ISDA that the market driven growth of credit derivatives has
substantially improved the transparency of corporate bond markets.             While
transparency in the credit derivatives market may not be optimal we do not see this as
reason to mandate transparency.

We understand that the Commission currently reviews UCITS separately. Here, we
only make the general remark that our view that differential regulation runs the risk of
moving liquidity from more regulated to less regulated markets also holds with
respect to the exclusion of UCITS from the intended scope of the Commission’s
report on non-equity market transparency.

We note that in the list of instruments on which you propose to focus, supranationals
have disappeared from those listed in 3.2 above. We suggest you include them with
government bonds together with government guaranteed agencies. Non-guaranteed
agencies should be included in the corporate category.

Question 5: To what extent do you consider there to be:
a. observable or demonstrable problems with respect to the possible policy
rationales
for mandatory transparency identified above in relation to one or more of the
instrument markets under review?
b. evidence that mandatory pre- or post-trade transparency would solve any of those
problems?

We find question 5 a. a little hard to understand. Do you mean, do we find problems
with the policy rationales or do we find problems in one or more of the instrument
markets under review?

As indicated in our paper AMTE believes that in existing MiFID level 2 Directive
needs to be implemented and digested by the market before it is possible to make a
full assessment of the benefits of mandatory pre- or post-trade transparency. In
general AMTE believes that pre-trade transparency is adequate or that efforts to
impose increased pre-trade transparency will be hard to implement and have possible
negative consequences. Best execution requirements of MiFID are likely to increase
the use of MTFs and regulated exchanges for the purpose of best execution.

End of day post-trade transparency is likely to best match the competing objectives of
dealers and investors as well as other actors. Where possible transparency
requirement should be introduced on a self regulatory basis with the Commission, if it
desires, maintaining a watching brief.




                                                                                     50
Question 6: To what extent could recent and upcoming technological and market
developments in relation to the instrument markets under review:
a. contribute to a relatively inexpensive extension of mandatory transparency?
b. render mandatory transparency unnecessary?

See our response to Question 3.

Question 7: To what extent are non-equity financial instruments different from
equities so that lower levels of mandatory transparency in those markets may be
justified?

Most academic work on markets is based on the broker market model rather than the
dealer market model. Bond markets naturally fall into the dealer market category
because of the greater ability of bank dealers to hedge and hold inventory. This
improves liquidity of the market and reduces the cost of capital for issuers. In equity
markets dealers are less able to hold inventory and have less satisfactory hedging
instruments thus the market gravitates to the broker model where maximum pre-trade
transparency and the use of regulated exchanges is necessary to order to match buyers
and sellers simultaneously.

The ability to price cash bonds using interest rate and credit default swaps as well as a
spread to government bonds improves the ability of the market to determine and to
check prices. Thus, notwithstanding the vast quantity of bonds on offer, it is possible
to rank and compare them reliably by using government and swap curves as well as
CDS. Similar comparisons for equity markets are less reliable.

Question 8: What data sources do you consider relevant to the issues you have
raised (if appropriate, cross-refer to your answers below)? Would you or your
organisation be
prepared to provide any relevant data if necessary?

iBoxx data is useful though prices are not traded prices and data series are relatively
short. The indices are limited to more liquid instruments.

Question 9: Are there academic or institutional papers or ongoing work that should
be
considered in preparing the Report not included in our bibliography?

ECB Occasional Paper “Implications for liquidity from innovation and transparency
in the European Corporate bond market”; August 2006. We feel our own research on
the distribution of daily returns contained in Appendix 3 of the attached paper might
usefully be followed up.


                                                                                      51
Question 10: What conclusions do you draw from the existing academic debate and
the work being conducted by other interested parties?

The impact of mandatory real time post trade-transparency on European bond markets
is unproven. We accept that there is a market failure in the absence of adequate
trading data on European bond markets to be used for valuing portfolios, determining
liquidity and volatility for regulatory purposes and for academic research. We believe
that this can be resolved adequately without significant risk to market liquidity by
introduction of an end of day reporting system with data available publicly and at
reasonable cost. Data on individual bonds should be made available (at a proportional
and reasonable cost) to allow investors to check best execution ex-post. We would
recommend that such a system of post-trade transparency be established on a non-
mandatory basis and should be allowed to evolve with the market. The Commission
could maintain a watching brief on this process and in particular review studies of
market behaviour based on this data and conduct its own studies.

Of more significance is the absence of market wide information on aggregate
positions. This is discussed in the ECB Occasional Paper mentioned in our response
to question 9. We believe that European regulatory bodies working together with
trade associations and individual market players should review how data could be
consolidated and published.

Much academic research is based on the broker market model common in equity
markets. We argue that this model is more applicable to equity markets where dealers
have less ability to take on inventory and where maximum pre-trade transparency is
necessary to match buyers and sellers simultaneously. Bond markets benefit from
the ability of banks to warehouse and hedge positions. To benefit from this, dealer
markets with lower levels of pre-trade transparency are necessary. To a significant
extent, there has been a cross over in market models with larger equity trades
executed OTC with dealers and smaller bond trades gravitating to MTFs and
regulated exchanges with greater degrees of transparency.

Question 11: In your view, how applicable is the academic or institutional literature
concerning transparency in the cash equities markets to the present discussion?

There has been little academic research based on European bond market data.

We would like to see research comparing bid/offer spreads with distribution of short
term returns in European and US markets.

We would also like to see more research on the impact of regulation in creating
barriers to entry. The relatively heavily regulated US market has significantly fewer
market makers than equivalent European markets. In particular, we would like to see

                                                                                   52
research on the impact of regulation on the ability of smaller specialist market makers
operating especially in the less liquid segments of the market.

A comparison of corporate new issue fees in Europe and the US would also be
instructive.

In addition, research into the actual price paid for bonds by retail investors in the US
using the US broker system should be compared with the prices actually paid by retail
investors, principally through bank outlets, in the European markets.

Question 12: What similarities, and what differences, are there between US and EU
markets that should be borne in mind when seeking to draw inferences from the
TRACE experience in the US?

In general, the US market is less competitive than the European markets. There are
significantly fewer dealers in the US market probably because of greater regulatory
intervention leading to barriers to entry.

It seems probable that retail clients in the US obtain substantially worse prices relative
to wholesale market prices than do retail clients in Europe.

Question 13: To the extent that you have identified problems or believe that others
might do so, do you agree that only EU-level action would be appropriate in the
present case?

We strongly believe that integration of the EU markets is the most important
contributor to liquidity and lower capital costs. Any independent action by individual
EU regulators is likely to harm the cause of market integration.

Question 14: If you have identified problems or believe that others might do so, to
what extent do you consider those problems would disappear as a natural product
of market evolution in the short-to-medium term?

The development of interest rate and credit derivative markets have improved
dramatically the transparency and price formation capacity of the market, as well as
risk management capabilities. Technological development has substantially increased
transparency. These developments have been unimpeded by regulatory intervention
and present a model for future developments of the markets.




                                                                                       53
Question 15: In respect of both pre- and post-trade transparency, are the four
options the right ones to consider, and in particular should other options be
considered?

As discussed in our submission, we believe that there is a case for regulators to
compile and to provide the market with more information on systemic risk based on
estimation of aggregate risk positions built up by the market. Information on net
positions by trading venue may assist in improving search liquidity in this respect.
The technical difficulties of deriving an aggregate risk position not only for a given
trading venue but for the market as a whole should be carefully analysed by the
Commission and be consulted with the relevant market participants.

Question 16: Would you, in light of your answers to the other questions, favour any
of the four options in relation to pre- and post-trade transparency (or another
option you might propose for consideration) in respect of transactions in any of:
  cash government bonds;
  cash investment-grade corporate bonds;
  cash high-yield corporate bonds;
  asset-backed securities;
  credit default swaps, interest rate swaps and bond futures; or
  any other financial instrument you consider relevant?

Differential regulation always runs the risk of moving liquidity from more regulated
to less regulated markets. To the extent that regulation or self-regulation is deemed
necessary we recommend that it be applied equally to all cash bond markets
regardless of the credit quality of the issuer. However, we recommend that a
differentiation be considered according to the type of instrument (cf. our response to
question 2) taking into account the standardisation/commoditisation and liquidity of
the relevant debt instrument.

We do not recommend any mandatory pre-trade transparency for bond markets.
MTFs and regulated exchanges may impose transparency requirements and we
believe that the best execution requirements of MiFID will effectively force pre-trade
transparency for best execution clients. The improvement in pre-trade transparency in
recent years as a result of technological innovation makes mandatory pre-trade
transparency unnecessary.

We recommend post-trade reporting at the end of the trading day. Trades in excess of
EUR 5 million should be reported as EUR 5 m+ with full size disclosed 24 hours
later. This should be implemented on a self-regulatory basis. OTC derivatives
markets have different problems particularly in reporting post-trade. A series of
periodic published market fixings for the more liquid OTC derivative instruments


                                                                                   54
would assist investors and regulators in evaluating cash bond markets. Again, this
should be instituted on a self-regulatory basis.
The Commission may if it feels appropriate maintain a watching brief on this self-
regulatory action. Improved data availability on European bond markets through
post-trade publication should allow more and better research on the liquidity and
transparency trade off.

We make no recommendations on commodity and other financial markets.




                                                                               55
Appendix 5: AMTE Working Group “Bond Market
      Transparency in Europe” Members




                                              56
AMTE Working Group “Bond Market Transparency in Europe” Members

   •   Aurélie Bedouet, AMTE
   •   Charles Berman, Citigroup
   •   Valérie Blanchin, AMTE
   •   Eric Brard, Société Générale Asset Management
   •   Won Choi, Citigroup
   •   David Clark, EIB, rapporteur of the AMTE Working Group
   •   Marie-Hélène Crétu, Euronext
   •   Frank Czichowski, KfW, co-chair of the AMTE Working Group
   •   Bertrand de Mazières, EIB, co-chair of the AMTE Working Group
   •   Stefan Ericsson, ABN Amro
   •   John Fleming, Credit Suisse
   •   Xavier Gallet, Veolia Environnement
   •   James Garvey, Goldman Sachs,
   •   Françoise Guillaume, Société Générale Asset Management
   •   Eric Haza, Veolia Environnement
   •   Baptiste Janiaud, Veolia Environnement
   •   Kenneth G. Lay, The World Bank
   •   Jochen Leubner, KfW, rapporteur of the AMTE Working Group
   •   Matthieu Louanges, Pimco
   •   Philippe Musette, Calyon
   •   David Newman, Citigroup
   •   Xavier Pujos, BNP Paribas
   •   Lisa Rabbe, Goldman Sachs
   •   Philippe Rakotovao, MTS Group
   •   Rodolphe Sahel, Société Générale
   •   François Systermans, BNP Paribas
   •   Daniel Trinder, Goldman Sachs
   •   Ivan Zelenko, The World Bank


Invitees:

   • Emmanuelle Assouan, Banque de France
   • Marco Laganà, European Central Bank




                                                                       57
Appendix 6: AMTE Board members and list of AMTE
              institution members




                                                  58
AMTE Board members


Chairman:
René Karsenti
Executive President, ICMA

Treasurer:
Jean-Pierre Mustier
Chief Executive Officer, Société Générale Corporate and Investment Banking

Secretary:
Jean-François Boulier
Head of Euro Fixed Income and Credits, CAAM (Crédit Agricole Asset
Management)

Board members:

Charles Berman
Managing Director, Co-Head of Fixed Income Capital Markets, Europe, Citigroup

Benoît Coeuré
Deputy Chief Executive, Agence France Trésor

Joe Dryer
Managing Director, Co-Head of Corporate Finance and Origination, DrKW Germany
of Dresdner Bank AG

Heinz W. Fesser
Global Head of Fixed Income and Managing Director, DWS Investments

James Garvey
Partner Managing Director of Investment Grade Coverage, Goldman Sachs

Robert B. Gray
Chairman, Debt Finance & Advisory, HSBC Bank plc

Matthieu Louanges
Generalist Senior Portfolio Manager, PIMCO

Bertrand de Mazières
Director General – Finance, EIB


                                                                                59
Gerhard Schleif
Managing Director, Bundesrepublik Deutschland Finanzagentur GmbH

Secretary General: Valérie Blanchin


List of AMTE institution members

• ABN Amro                               • Eurex
• AFT (Agence France Trésor)             • Euroclear
• AFTE (French Association of            • Eurohypo
Corporate Treasurers)                    • Euronext
• AFTI                                   • Goldman Sachs
• Axa Investment Managers                • Hellenic Republic – Public Debt
• Bank of America                        Management Agency
• Banque de France                       • Hengeler Mueller
• Barclays Bank                          • HSBC
• Bloomberg                              • HVB
• BNP Paribas                            • ICAP Electronic Broking
• BNP Paribas Asset Management           • IXIS Corporate & Investment
• Bundesrepublik Deutschland             Bank
Finanzagentur                            • JP Morgan
• CAAM (Crédit Agricole Asset            • KfW
Management)                              • LCH.Clearnet
• CADES (Social Security Debt            • Lehman Brothers
Repayment fund)                          • MarketAxess
• Calyon                                 • Merrill Lynch
• CdF                                    • Moody’s
• CIF Euromortgage                       • Morgan Stanley
• Citigroup                              • MTS Group
• CNO (Bond Standardisation              • Natexis Banques Populaires
Committee)                               • Nomura
• CME (Chicago Mercantile                • PIMCO
Exchange)                                • S.G. Asset Management
• Crédit Foncier de France               • Société Générale
• Credit Suisse                          • The Royal Bank of Scotland
• DEPFA Bank                             • The World Bank
• Deutsche Bank                          • TradeWeb
• Dexia Crédit Local                     • UBM
• Dresdner Bank                          • UBS
• DWS Investments                        • Veolia Environnement
• EIB (European Investment Bank)         • Viel & Cie
• ETC Pollak Prebon

                                                                             60

								
To top