Preliminary version that may be amended in the light of the Discussion

The Euro Bond Market: Developments and Implications for

Monetary Policy
By Graham Bishop
Paper for European Central Bank Seminar, May 5/6 2000

Note: This paper is delivered in my new role as an independent consultant

GrahamBishop.com as well as continuing as an Adviser to Salomon Smith Barney.
Nonetheless, the views expressed are the authors own and are not necessarily
shared by Salomon Smith Barney or any of its affiliates.

The purpose of this paper is to consider the developments in the euro bond market since
its inception and highlight any trends that might impact the operation of monetary
policy. The paper attempts to project the observed trends forward to the logical
conclusion of the process. The analysis is split into three parts:
1. Consequences of introducing the euro: Technical, legal and other changes.
2. Developments in the euro denominated markets since inception.
3. The potential impact of the future markets on the operation of monetary policy.

Conclusions:
Without making extreme assumptions, it is not clear that the initial changes in
the euro bond market could yet sum to an amount that would have a
significant impact on broad money. Crucially, monetary policy is not
mechanistically determined by monetary growth, so the risk of a significant
distortion of policy seems modest at least in the brief period of the euro
bond markets existence so far.

However, such a large bond market may eventually make it easy for savers to
hold the ECB and politicians to account on the outlook for both inflation and
public finance via, effectively, a rolling referendum.

PART I: Consequences of introducing the euro: Technical, legal and

other changes.
The European financial services industry as a whole is changing rapidly under the impact
of several major driving forces. Most obvious is EMU itself but liberalisation is
proceeding apace and technology is advancing into an exponential change phase that may
well allow a dramatic dis-intermediation of many parts of the financial system. The first
part of the paper gives an overview of the specific changes that seem most likely to
impact the market structure within which the ECB will have to implement its monetary
policy. Naturally, the first set of changes were those that were required to operate a
monetary union of some 300 million people.

1
Simultaneously, several other fundamental factors have been at work in the background
and the full implications of these will only be apparent in the medium term. Nonetheless,
the consequences have the potential to have a very powerful effect on the implementation
of monetary policy so they should be carefully monitored.

Technical Steps by the ECB

As part of the preparations for EMU, the ECB took a number of technical steps so that it
could implement its monetary policy efficiently. Some of these developments might have
occurred naturally in due course, as the technology became so readily available that cost
reduction pressures would have pushed market participants to adopt these measures
anyway. Nonetheless, EMU certainly accelerated the following key steps:

Real time cross--border payments were essential. To create a single pan-Euroland

money market, it is necessary for market participants to be able to arbitrage any
geographical price discrepancies in money. The perfection of that arbitrage will
depend on the efficiency of payments systems so very large volumes of money would
need to be shifted rapidly. If the central banks wished to avoid potentially massive
credit exposure to the payment banks, then a real-time system was the logical choice
and technology had just made it possible. So the TARGET system was born.
Real time securities settlement was another necessity, once the choice of avoiding
central bank credit exposure had been made. Execution of monetary policy via
securities transactions required the ability to settle very large volumes of securities
flows instantly. Moreover, the need for collateral movement as part of the payment
system pointed in the same direction. Once the ECB had decided it would require
these settlement systems to be built for the limited purposes of monetary policy
implementation, then the starting gun had been fired on a major re-structuring of the
entire settlement chain. Moreover, the ECBs minimum standards for depositaries
also highlighted the need for some action. With this limited process underway, it was
inevitable that market participants would apply the risk reduction and cost saving in
all sectors of their activity.
A general broadening in the use of collateral was a natural response to the
development of systems for dealing with the ECB. Its definitions of Tier One (high
credit quality, marketable debt instruments) and Tier Two (national instruments)
seem wide enough to cover new securities that may be developed as market
participants evolve new products to appeal to pan-EU investors. Nonetheless, the
importance of the ECBs eligibility criteria should not be over-estimated. Total
collateral held by the ECB was ELOOLRQRXWRIDSRRORIHOLJLEOHDVVHWVVWDQGLQJ
at ELOOLRQLQMarch, 2000. Salomon Smith Barney calculates the size of
markets that are readily accessible to institutional investors by virtue of a minimum
issue size of PLOOLRQ7KDWWRWDOZDVYLUWXDOO\ ELOOLRQDWWKHHQGRI0DUFK
2000. Given the wide eligibility criteria, it would be surprising if a newly issued
security were not ECB eligible and there seems to be hardly any price improvement
at issue or in secondary trading, even in the repo market itself for eligibility.

2
 Improving the legal certainty of collateral is an obvious step and the Giovaninni
Group issued a report on this in October 1999. In particular, it urged Member States
to apply a broad interpretation of Article 9(2) of the Settlement Finality Directive so
that the protection of legal certainty was not unnecessarily narrowly defined to
include merely the ECB and payments systems. Instead, they should eliminate legal
risk for all participants in EU settlements systems.
Retail cross-border payment systems fall into the ECBs basic mandate of
overseeing the payments systems. Retail systems were subject to a new Directive in
late 1999, but that may not be enough to create cross-border systems that are as
efficient (fast and cheap) as domestic by 2002. Citizens are unlikely to be impressed
if they cannot move the actual single money around to take advantage of price
discrepancies. Indeed they may wonder what at a technical level - the purpose of a
single currency was. There may be serious political implications from any consequent
loss of popular enthusiasm for Europe at a time when IGC ratification will be on
the agenda to enable enlargement to occur.

Recasting the market infrastructure

The ECBs 1997 framework for monetary policy may have implied a major shake-up in
the financial system but action is now underway. The first quarter of 2000 witnessed a
renewed upsurge in announcements, and actions, to recast the markets infrastructure,
whether in trading, or clearing and settlement. This realignment represents a progressive
creation of the building blocks that will constitute the future European securities
market(s).

The impact of Internet equity brokers in the US is well documented and that has spread
rapidly to the creation of Electronic Crossing Networks (ECNs). But the proposed new
partnership between the Pacific Stock Exchange and Archipelago takes the process
another step along the road because this would become a fully regulated, but electronic,
stock exchange. European equity exchanges seem to be some way behind this but the
vision is shared and the chances are that the same process will flow over into bonds in
due course.

Certainly, the bold move by the French, Belgian and Dutch Stock Exchanges to merge
into Euronext offers competition to the demutualising London Stock Exchange and the
about-to-be-quoted Deutsche Borse. At the same time, Euroclear and Sicovam announced
that they would merge so that they can provide Euronext with, straight-through
processing services and consolidated reporting, from trade to netting to
settlementThese services are also available to other stock exchanges and trading
platforms. Recently, the London Clearing House (LCH) and Clearnet ParisBourses
settlement system announced a merger to provide a single European clearing house
with a central counterparty. Again, press reports stated their willingness to talk to others.

3
Other depositaries are also gearing up. Cedel and Deutsche Borse Clearing cemented their
merger as Clearstream and promptly announced that they are looking for further
consolidation. Outside Euroland, the UKs Crest has forged closer links with its counterpart
in the Swiss market SIS and has also linked to the US depositary DTC to
provide individuals with easy settlement of leading US shares.

EuroMTS continued to expand its electronic inter-dealer network and, within a year of
starting operations, has achieved an estimated 30%-40% market share. This year, Finnish
government bonds extended its trading list to nine Euroland governments. It started trading
repos in October 1999 and the LCH became a central counterparty in December to
improve security further. During the second quarter of 2000, Brokertec should start
competing with EuroMTS in euros, as well as offering a similar service to the US Treasury
market.

Bond trading with customers has not been immune to the downstream changes: several
European bond issues were made over the Internet in the first quarter. Initially, this may
just be information but the Finnish Government issued a bond where investors could
place orders on its dealers websites. However, the next wave may already be in sight as
several leading dealers including Salomon Smith Barney launched Syndicate.Hub
to broadcast details of offerings. This gives access to electronic roadshows and
prospectuses, as well as allowing investors to give an indication of their interest.
Effectively, it will be a portal into the lead managers own system such as Salomon
Smith Barneys SSB Syndicate Direct.

Liberalisation by the politicians

Last year, European bond markets acquired a momentum for change driven by investors
thirst for yield and diversification. The rapid pace of change in the physical infrastructure
opens the way to even more transformation. But that process is likely to have limits that
reflect the legal infrastructure.

Finance Ministers and the European Commission recognised this possibility back in 1998
and launched the Financial Services Action Plan (FSAP) to identify, and then remove, the
barriers to a single capital market. The process is gathering pace and the March Summit of
EU Heads of Government in Lisbon gave liberalisation further impetus at what was
called the dot.com Summit.

The Summit produced two particular, and strong, conclusions that are relevant to the
capital markets:

1. That pending legislation should be adopted during 2000 on the legal framework for e-
commerce, e-money, and the distance selling of financial services.
Also, the Commission and the Council to consider how to promote consumer
confidence in electronic commerce.

4
2. To accelerate completion of the internal market for financial services, steps should
be taken: to set a tight timetable so that the Financial Services Action Plan is
implemented [authors emphasis] by 2005by means of a single passport for
issuers eliminating barriers to investments in pension funds; promoting further
integration and better functioning of government bond markets through greater
consultation and transparency on debt issuing calendars, techniques and instruments;
enhancing the comparability of companies financial statements

The Heads of Government have now set themselves deadlines so that their actions can be
measured against the rhetoric. That should be a powerful discipline otherwise it will
open the EU as a whole to the charge that there is more talk of reform than action!
As the Summit Communiqu put it, Efficient and transparent financial markets foster
growth and employment by better allocation of capital and reducing its costs. For the
FSAP, the commitment is to have it implemented by 2005 which plainly means that
the relevant national laws will allow all the proposed actions to be legally valid throughout
the EU after 1 January 2005. The revolutionary changes that are underway in the
market infrastructure should have come to fruition on a similar time scale, so the
scene is set for a transformation of the European financial services industry.
Undoubtedly, that will influence both the technical operation, and transmission
channels, of monetary policy.

Regulatory Change

The Basel Supervisors issued a discussion document for response by March 2000 so that
they can issue definitive proposals later in the year. These will be incorporated into EU
law by 2002, according to the Financial Services Action Plan. That future timetable
immediately impacted market activity in structuring deals that may mature after the
implementation date, thus posing capital uncertainties to banks that hold the bonds or
are securitising assets.

The implications for the bond markets are mixed as the main changes are proposed to the
capital required for claims on the corporate sector. For those top-rated companies (AAA to
AA), the capital weighting would fall from 100% to 20% but these companies are
precisely those that have ready access to the bond market already. Moreover, only a modest
number of European corporations are rated by Moodys in this higher category. However,
claims on companies that are well below investment grade (B or lower) would attract a
150% capital charge, so this could herald an accelerating build-up of high yield bonds in
Europe. This would fulfil the public policy objective of increasing the flow of capital to the
riskier companies, which are often those that are increasing employment more rapidly.
However, it is conceivable that banks will re-price loans to companies that are close to that
threshold to take account of the risk that a credit downgrade would suddenly trigger a 50%
increase in the capital requirement with a corresponding cut in the banks return on its
shareholders capital.

5
The Basel Committee has undertaken to study further the treatment of market risk where
some banks may try to arbitrage regulatory capital for instance, by holding bonds on their
trading book as a loan substitute, which when fully hedged against market risk and
subject to certain conditions may attract a charge of only about 20%.

Demographic Effects and Retirement Savings

Europe is beginning to grapple with the financial consequences of an ageing population.

But movements towards fully funded, Second Pillar systems of providing retirement
income are erratic. However, the demographic trends in the population are so well know
that citizens who can afford to make provision for their eventual retirement are likely to
do so long before it is mandatory. Indeed, the upsurge in mutual fund sales across Europe
suggests that this process is already underway.

Whatever the political choice between encouraging Defined Benefit (DB) plans or
Defined Contributions (DC) systems, there may well be a rough and ready matching of
desired assets towards the aggregate liability of EU pensioners-to-be. The UK is already a
long way down the road of funded pensions and there may be some useful lessons to be
learnt about the consequences of the huge investment flows that occur. Moreover, in a
price stable world, citizens will have to come to terms with the unwelcome facts of
nominal investment returns that seem very low by the inflation-related standards of recent
decades. The target pool of savings at retirement will have to be much higher and the
thirst for absolute yield is likely to be a driving force in the years ahead. In combination
with the technological capacity to cut dealing costs, this could well be the driving
force that securitises credit in Europe. This could have a profound influence on
the implementation of monetary policy.

The Lessons from the UK

The early stages of this process are illustrated by the problem facing UK pension funds
with assets around 850 billion or 95% of GDP - the Minimum Funding Requirement
(MFR). This was introduced by the Pensions Act of 1995 for actuarial valuations after 6
April 1997 - and these must be up-dated every three years. So April 2000 marks a new
phase of maturity for this process. Assets must be not less than the accrued liabilities
from now onwards and the liabilities are calculated using the long gilt yield as the
discount rate for (1) retired lives and (2) a progressive blend of "equity returns and that
long gilt yield" once a scheme member is within 10 years of retirement.

6
The consequences compound: Increased life expectancy plus rising numbers moving into
the "less than 10 years" category gives a double hit in shifting the liability calculation
onto the gilt yield basis. As Trustees are legally required to maintain sufficient assets, it
seems reasonable to expect them to minimise the risk of breaching the Act, so they may
well pay increasing attention to the nature of the liability that they are matching. Just as
the banking community sought to minimise its regulatory risk versus the Basel capital
standards, UK pension fund trustees are likely to take the same path of minimising
personal liability.

The implications of the rising tide of retirement savings can be split into two. First, there
is the sheer quantum of funds that will flow from the rising dependency ratio: The over-
65s are set to rise from around 20% of the working age population to about 40% in the
next 40 years. But, secondly, the driving force for this bulge is increased longevity
and that will determine the nature of the assets that the ageing population will wish
to hold.

The longevity problem has been accelerating for decades and Figure 1 shows the trends
in the life expectancy of a 65-year-old male in the UK. The problem may be even more
acute for the pensions and life insurance industry as the life industry - the source of these
figures - seems to attract customers with a considerably better life expectancy than the
population as a whole. There is no reason to think that the aggregate experience of the
EU will be much different.

Figure 1: Changing Life Expectancy at Age 65, UK Male

18

16

14

12

10

0
1920 1925 1930 1950 1960 1970 1980 1990
WholeUK Population AssuredLives

Source: Continuous Mortality Investigation Bureau (CMIB) of the Faculty & Institute of
Actuaries.

7
However, this chart does not show the complete story as the effective retirement age is
often much lower than 65. The Continuous Mortality Investigation Bureau of the Faculty
and Institute of Actuaries published new tables - with data for 1992 - in July 1999. A man
retiring at 60 is currently expected to live for 23.75 years - up 13% from the expectation
in 1980. A 55-year-old retiree can expect another 28.83 years - up only 9% from that
earlier forecast. As medical science advances, albeit at great expense to the public purse,
these survival rates are expected by the government actuaries department to improve
further - for a 65 year old in 2040 by about 8% for the whole population. The survival
rate of insured lives is expected to rise by about 13%.

Demand for Securities

The MFR regulations impose a detailed set of asset/liability matching rules on UK

pension funds. So the demographic pattern of the liabilities should define the asset mix.
However, these funds have 71% of their assets in global equities but only 16% in global
bonds - according to CAPS, a leading fund measurement service. But, 38% of members
are pensioners and only 62% active/deferred members - where the liability would be
based on equity returns (Source: National Association of Pension Funds, 1998 Survey).
So there is a significant asset/ liability mismatch that this author has guesstimated at 10%,
though it is almost certainly much higher.

There is a risk of a further twist to this mismatch. If employers conclude that the costs
and responsibilities of running Defined Benefit Plans are too high, then - logically - they
will shut existing schemes and only offer Defined Contribution Plans to new members.
Thus existing DB schemes, and they hold the vast majority of the current 850 billion of
assets, will age progressively until, eventually, ALL the then-remaining assets have to
back liabilities calculated on a gilt yield basis. During the decade when the final segment
of members move through the ten years to retirement category, there would be a
massive requirement for fixed income securities. Regrettably, it appears that there is not
any aggregate data that would enable the scale and timing of this problem to be analysed.

The Implications for Euroland may be very important. One way or another, there may
be a structural, and massive, increase in the supply of capital with a desired term structure
that is more related to expected mortality than conventional economic analysis about
expected inflation. Indeed, real returns on long dated UK Index-Linked Gilts are now
down to around half the comparable yields in France or the US. Again, the weight (and
regulatory bias) of investible money is the most likely reason especially when the
Government is running a large budget surplus.

As a result the UK yield curve is extraordinarily distorted with 30-year bonds yielding
nearly 100 Bp. (and recently much more) less than 10-year bonds (see Figure 2).
Currently long Gilts yield over 100Bp. less than corresponding German bonds a
margin that has declined from 140 Bp. in recent months. Since the US government
announced that it would buy back some of its debt, the US yield curve has also inverted.

8
Figure 2: Par Yield Curves for Government Bonds in US Dollars, Sterling and
Euros (Percent)- as at 25 April, 2000

%
7

6.5

5.5

4.5

3.5

3
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
Year USD GBP EUR

Source: Salomon Smith Barney

The governments of Euroland have been moving towards budgetary balance but are not
yet facing the problems of surplus. Also their debt managers have taken advantage of the
decline in yields triggered by EMU entry to extend the average life of debt. The Euroland
central government bonds in the Salomon Smith Barney Performance Indexes now have
an average life of 6-7 years. But this is still dramatically shorter than the 10 20 years
that ageing savers are likely to be seeking.

The combination of sound public finance - thus reduced demand for capital - and a
mismatch with the desired maturity structure of the suppliers of capital could yet
have consequences for the operation of monetary policy due to the impact on the
shape of the yield curve. The mechanism for intermediating the capital flows from
the savers to the borrowers may sharpen the impact. Will the process be based on
the existing institutional structure? Or will it alter, perhaps radically, as a
consequence of the technical, legal and technological changes now underway?

9
Management of Government Debt

The biggest single sector in the fixed income markets is government debt so its
management sets a powerful example (see later discussion on recent developments). The
creation of a single bond market in euro has focussed much attention on the yield spreads
between different sovereign states. As a result there has been much debate about the need
for a single Debt Office so as to create a benchmark yield curve. This subject seems
entirely technical but, in reality, it goes to the heart of the political relationship amongst
EU Member States because the ability to borrow is a critical ingredient in politicians
perception of sovereignty. Borrowing balances the spending commitments given to the
electorate with the level of taxes to be imposed.

Any proposal to change the existing borrowing arrangements needs to be subjected to a

careful cost-benefit analysis where the political dimension is a very significant if
unquantifiable item. Moreover, the first question to be answered is whether there is a
general demand from borrowers or investors for a change in the current arrangements.
It is not clear that investors are seeking anything more than incremental improvements
that are likely to be delivered naturally by market forces over the next few years.

What benefits might accrue?

Borrowers point to the existence of yield spreads that are most noticeable at the longer
end of the curve recently ranging up to about 30 basis points for 10-year bonds.
Naturally and properly, the debt managers of those States at the top of this range see it as
their duty to seek a reduction. However, they are constrained by the credit outlook for
their State and can focus only on the technical aspects that are under their control.
If spreads could be induced to narrow only through a decline in the yields paid by the
high-spread States, then the savings on an EU-average debt, of say 70% of GDP, would
still be modest. Figure 6 below shows that only about 52% of Gross General Government
Debt (The Maastricht Treaty definition) is financed via bonds issued by the central
government that are accessible to major institutional investors. Salomon Smith Barney
measures this by the definition used for its well-known family of Performance Indexes. If
as much as half this bond debt were subject to the 10-year spread, then its disappearance
would eventually cut borrowing costs by perhaps 0.05% of GDP. Whilst it is right that
every opportunity be taken to minimise public expenditure, such a relatively modest
benefit implies that any countervailing disadvantages should be weighed up carefully

Disadvantages

These can be split into political and technical issues. These disadvantages need to be
examined under the most stressful conditions to see whether problems that might seem
remote and minor under normal market conditions could have the potential to create
serious strains that could even threaten the continued existence of the EU.

10
 Political disadvantages: These stem from the possibility that the discipline imposed
on finance ministers by the markets discipline of worsening credit spreads could be
blunted by reducing the amount of debt exposed to borrower-specific market forces.
In effect, this would risk negating the impact of the no bail out rule. Many if not
most - market participants already regard this rule as illusory and any steps that re-
inforced such an analysis would raise the risk that its application would be such a
shock that it could de-stabilise the financial system. Any system of cross-guarantees
increases the risk that investors will assume that it will be stretched further in crisis.

Whilst a State that becomes subject to the Excessive Deficit Procedure, theoretically,
could be expelled from the joint borrowing arrangements, a stress-test analysis points
to the risk of such an action de-stabilising the State at precisely the most sensitive
moment. This is the key reason why many market participants doubt that the
Excessive Deficit Procedure would ever be rigorously applied in the first place let
alone increasing the cost of the sanctions by raising that States cost of borrowing.

Technical disadvantages: There is a significant risk that any narrowing of spreads

would be achieved at the cost of an increase in absolute yields for the most highly
regarded borrowers. If those governments wished to raise their contribution to the EU
generally, they might choose to use the money in a different way. Without that wish,
it is not clear why they would participate at all.

Any improvement in the absolute yields paid by the higher-spread States that is not at
the expense of higher yields for others is only likely to come about from improved
liquidity. Thus the requirement would be for single bond issues of about ELOOLRQ
matching the size of the typical Big-5 issuer. However, these States account for
nearly 80% of the bonds outstanding. If only say 20% of the bonds of the Small-6
were put into this scheme, then the ELOOLRQZRXOGKDYHWREHVSOLWLQWRDERXW
bonds of comparable size and this might take many years to build up, given the
pattern of maturities and other constraints.

The concept of a multi-issuer futures contract for these States needs to be examined
critically as the disappointing performance of such contracts so far does not bode
well. Liquidity of a futures contract is all-important and self-reinforcing. That is
obvious for dealers hedging their positions but speculators are interested in market
movements rather than the volatility of spreads between markets. Indeed, they are
likely to be concerned that unexpected risks such as sudden political developments
could shift this proposed contract to one that is dominated by a sudden change in
the cheapest-to-deliver bond.

11
 It is not clear that these yield spreads are susceptible to elimination to the extent that
bonds are used for collateral purposes and thus become special in repo. Tax effects
can also be important in the context of spreads of this size. Auction concessions may
be significant at times and rare but small issuers could well find that their rarity
operates to their advantage because investors are seeking a well- diversified portfolio.

Improvements that investors require are easily summarised transparency of issuing

timetables, greater ease of settlement and enhanced liquidity. Market forces are already
delivering much of these demands:
Auction timetables by the bigger issuers are on the way to becoming set by
convention as say France and Italy set out their schedules a year ahead. Other States
could follow suit.
Ease of settlement is a major factor in the current upheaval in the European
depository industry and the next few years are likely to see major improvements
delivered.
In recent years, many States have realised the need for large liquid issues and the first
year of the euro has witnessed several sets of exchange offers to enhance liquidity.

The Conclusions of the Lisbon Summit in March appeared to endorse the idea that these
more limited needs should be met in the near future and certainly by 2005.

PART II: Developments in the euro denominated markets since

inception.
The capital raised by private corporations and utilities direct from the financial markets
rose dramatically overtaking the central governments. Even in its first year, monetary
union seems to have triggered a significant shift towards the securitisation of credit
in Europe.

What does securitisation mean?

Securitisation is defined in the broadest sense, for this analysis. It means connecting
the suppliers of funds directly with the users via a market for securities, rather
than through an intermediary bank.
The term securitisation is often applied to the specific process of making small loans
perhaps on residential mortgages or even credit cards into bonds that can be issued on
the capital markets and purchased by large investment institutions. More generally, it can
include the process of governments transforming their non-marketable debts into highly
liquid bonds that command a lower interest rate and thus cost saving.

12
Figure 3. Central Government and Underwritten Bond Issuance, 1Q98-1Q00 (Euros in Billions)

300

250

200

150

100

50

0
1Q98 2Q98 3Q98 4Q98 1Q99 2Q99 3Q99 4Q99 1Q00
Private Corporate & Private Utility Private Finance Central Govt Issuance Underwritten issuance

Source: Capital Data Bondware; Salomon Smith Barney.

The decline in the financial demands of the public sector is the natural, and desirable,
result of the Maastricht Treatys insistence on sound public finance as a pre-condition for
EMU membership. So the trend to reduced government bond issuance may, eventually,
have much further to go. EU-11 gross general government deficit the Maastricht target
declined from 2.0% of GDP in 1997 to 1.7% in 1998 and 0.7% in 1999. Salomon Smith
Barney forecasts a further contraction to 0.5% in 2001.

For the financial markets, the key factor is the level of bond issues which reflects the
maturity structure of existing debt as well as the choice of financing vehicle. This might
be bank, money market or bond. Figure 3 demonstrates that the issuance of bonds has
been on a declining trend since we started collecting the data at the beginning of 1999.
Some governments notably Italy have been making efforts to lengthen the maturity of
their debt. But the sharp steepening of the yield curve since last Spring has reduced
enthusiasm for the moment.

13
Issuance of Underwritten Bonds

The rise in corporate issuance was a major trend in the new euro markets and has several
components:
The tsunami of M&A has produced a dramatic rise in the need for finance but it is not
clear which is the chicken and which the egg! After Olivetti managed to raise the
finance for its bid for Telecom Italia, the rules of the game have changed and now
Vodafones hostile bid for Mannesmann may result in a further wave of bond
issuance. Vodafones ELOOLRQEDQNIDFLOLW\ZDVUHSRUWHGWREHDGD\IDFLOLW\VR
much of it will need to be re-financed longer term. If European M&A activity
continues to run at anything like the ELOOLRQSDFHRIDQQRXQFHPHQWVLQ
and only 10% is re-financed in the bond market, then that would match the entire
corporate bond issuance in 1999.
In the investment grade sector, US companies became a major influence last year, in
part because of FAS 133 but also the desire by the finance companies to broaden their
investor base. Both these factors are likely to continue this year. Indeed, we believe
that FAS 133-related issuance could easily double.
Asset-backed securities (ABS) denominated in euro have just begun to scratch the
surface of potential issuance despite reaching ELOOLRQLQ7KHSRWHQWLDOVFDOH
of market that could result from the securitisation of European credit is most obvious
when considering the second largest component of this volume banks issuing
Collateralised Loan Obligations (CLOs) to reduce their need for regulatory capital.
The ELOOLRQLVVXHGLQRQO\DPRXQWHGWROHVVWKDQRIWKHEDQNLQJ
systems aggregate balance sheet.
Finally, the high yield market has developed very strongly as European institutional
investors have stepped up their credit research capabilities much faster than many
observers recognise. During the year, ELOOLRQZDVLVVXHGDQGPRUHWKDQKDOIRIWKDW
was in the fourth quarter alone driven by the telecom/cable companies. So the
capital markets are financing a vital part of Europes technological development.

The fact of this scale of issuance demonstrates that the European market for credit
products is now firmly established. Investors are unlikely to backtrack on their
evident enthusiasm for credit now that they have set up the portfolio management
facilities. Moreover, they will not wish to give up the extra yield even if rates rise
as demonstrated by the buoyancy of both asset-backed and high yield markets in the
fourth quarter.
Taking 1999 as a whole, the issuance of corporate bonds was roughly equal to the
budget deficits incurred by the EU-11 public sector. Yet that new debt capital in the
hands of the private sector appears to be oiling the wheels of an economic revolution
from the bottom up.

14
In the first quarter of 2000, Euro-denominated issuance rose 35% from the last quarter
of 1999, notably in the private bank and private corporate sectors. But that can be seen as
a rebound from the depressed levels ahead of the much-feared Y2K effect. Perhaps a
better comparison is with the second half of 1999 which showed a healthy 23% gain.
Measured against the boom in the first quarter of 1999, issuance this year fell by 10%.

Central Government also increased its underwritten issuance almost doubling its
volume from the first quarter last year, and a fivefold increase from the last quarter of
1999. Frequently, issuers have chosen to start the New Year with new lines of fungible
securities by launching an underwritten tranche to ensure a wide distribution and
immediate liquidity.

Figure 4. Issuance of All Euro-Denominated Bonds, 1998-1Q00 (Euros in Billions)

1Q00 1Q00 (%) 4Q99 (%) 3Q99 (%) 2Q99 (%) 1Q99 1999 (%) 1998
Amount (%) (%)
Sovereign/Government/Author 33 14.5 4.4 6.7 7.7 8.0 6.9 9.3
Central Government 28 12.5 3.0 5.6 5.3 7.0 5.5 7.2
Local Authority 1 0.6 0.6 0.2 1.2 0.5 0.6 0.6
State/Provincial Authority 3 1.4 0.8 0.9 1.2 0.5 0.8 1.5
Supranational 3 1.3 2.1 1.7 1.4 2.0 1.8 3.4
Supranational Institution 3 1.3 2.1 1.7 1.4 2.0 1.8 3.4
Private Corporate/Utility 25 10.9 9.8 17.1 15.1 10.4 13.1 5.2
Private Corporate 24 10.6 9.2 16.5 13.8 9.3 12.2 4.7
Private Utility 1 0.3 0.6 0.6 1.3 1.1 0.9 0.5
Public Corporate/Utility 9 3.8 3.4 3.2 3.2 2.1 2.9 2.4
Public Corporate 1 0.2 0.6 0.3 0.4 0.4 0.4 0.3
Public Utility 8 3.6 2.8 2.9 2.8 1.7 2.5 2.1
Finance 158 69.4 80.3 71.3 72.5 77.5 75.3 79.8
Private Bank 96 42.2 32.2 43.0 45.8 45.6 42.3 41.1
Public Bank 46 20.3 36.3 20.5 16.4 24.2 23.8 30.1
Private Finance (Other) 15 6.5 10.9 7.0 8.0 5.3 7.5 6.4
Public Finance (Other) 1 0.4 0.9 0.8 2.3 2.4 1.7 2.2
Total 100 100 100 100 100 100 100
Total Amount 228 168 202 215 251 836 590
Source: Capital DATA Bondware, April 2000.

The fourth quarter of 1999 may also prove to have been a watershed in the
international significance of the euro. The issuance of all types of GHQRPLQDWHG
bonds exceeded that of international dollar bonds by 81% as euro issuance surged by
52% from the depressed levels of late 1998 and dollar issuance fell by 20%. For the year
as a whole, LVVXDQFHH[FHHGHGLQWHUQDWLRQDOLVVXDQFHLQGROODUVE\DQGWKLVOHDG
was broadly maintained in the first quarter of 2000.

15
Despite the continued weakness of the euro versus the dollar, in the first quarter of 2000,
investors purchased 25% more euro-denominated bonds than international dollar issues.
However, that was a less pronounced margin than the 32% recorded in the whole of
1999. For such a fledgling currency, that performance should be seen as a major success
against the backdrop of a weak currency and rising bond yields.

As the vehicle of choice for private corporate borrowers, the dollar maintained its margin
of around a quarter.

Figure 5. Issuance of All Euro- and Dollar-Denominated Bonds, 1999-1Q00 (Euros in Billions)
1Q00 1999

All GHQRPLQDWHG Int US$- All GHQRPLQDWHG Int US$-denominated

Sovereign/Government/Authority 33 16 58 35
Supranational 3 10 15 25
Private Corporates/Utilities 25 32 110 135
Public Corporates/Utilities 9 4 24 13
Private Finance 111 52 416 220
Public Finance 47 68 213 206
Total 228 182 836 634
Memo: EU 6 Central Govt 121 437
Source: Capital DATA Bondware, April 2000.

Methodological Note: These comparative issuance tables are designed to capture the
flavour of the bond investment opportunities that are available to investors that have a
ready choice of currency because they are not legally constrained. Thus the analysis uses
the international, rather than domestic, component of dollar issuance. In Euroland, the
intention is to be as inclusive as possible, largely because the markets are moving
legally and technologically in that direction. Indeed, public borrowers have taken a lead
in Internet distribution. For this edition, we have expanded the definition of Government
issues to include all those that are underwritten, rather than auctioned, even if they are
classified as a domestic bond. The historic data has been adjusted correspondingly. In due
course, it seems likely that even these distinctions may become blurred.

Government bond markets

During 1999, perhaps one of the most significant developments was the success of
EuroMTS the electronic broking system developed from the platform used to trade
Italian Government bonds. Launched in April, it has expanded rapidly to cover 9 of
the Euro-11 government markets and we believe its market share has already reached
about 30-40% of inter-dealer volumes. This success was underlined by the December
decision to include Pfandbriefe as an initial step into the non-government market
but still targeted at the liquid component. The improved investor confidence in
liquidity and price transparency was demonstrated by a modest tightening of the
spread of 10-year Portuguese bonds after their benchmark bond was listed in
November.

16
 Moreover, EuroMTS has now announced a facility to arrange anonymous repos using
the London Clearing House as the central counter-party. These developments
underline the growing importance of the liquid benchmark issues and the associated
ability to create a secured money-market. A necessary component is the drive to
create legal certainty throughout the EU for this type of activity, highlighting the need
for a broad interpretation of the Finality Directive (see discussion in Part I).

Figure 6. Government Debt (Euros in Billions)

Bondsa as % of Market Value 10-Year Benchmark (Outstanding)

Country Maastricht Debt End-1998 Mar 00 Mar 00 Recent

Italy 33 544 21 23
Germany 39 522 20 20
France 58 501 24 23
Spain 54 201 18 13
Netherlands 65 163 12 11
Belgium 51 156 5 11
Memo:
EU-11 52 2,248
US 36b 1,848 10 22
Japan 33b 1,915 14 14
UK 49 396 14 17
a
Bonds as defined in SSMB World Govt Bond Index: fixed rate, over one-year remaining life. b US/Japan debt is OECD financial liabilities definition.
NOTE: The ratio of central government fixed-rate bonds to the general government sectors debt (as defined for the Maastricht Treaty) is designed to indicate
the degree of securitisation of public debt.
Sources: European Commission, OECD and Salomon Smith Barney.

Debt re-structuring was a continuous feature during 1999 as debt managers sought to
improve the liquidity of their issues. For example, the Dutch Treasury announced in
June a Debt re-structuring was a continuous feature during 1999 as debt managers
sought to ELOOLRQH[FKDQJHRIIHUprogramme to build up their liquid benchmarks
to a typical size of ELOOLRQ
Government bond markets continue to evolve rapidly as debt managers respond to the
new environment of low issuance and rising technology. Finland led the technological
way with an 11-year bond that was partly sold via the Internet.

Figure 7. Government Issuance for EMU 6 (Germany, France, Italy, Spain, Belgium and the Netherlands)
Issuance by Maturity
2-3 yr 4-5 yr 10 yr 15 yr 30 yr Total Gross Issuance
1Q99 49.8 33 53.5 1.7 15.9 154.0
2Q99 41.3 30.2 35.2 0.8 5.1 112.7
3Q99 31.2 19.8 42.1 1.3 5.1 99.7
4Q99 24.5 14.5 26.3 0.6 4.7 70.5
Total 1999 146.8 97.6 157.2 4.4 30.8 436.7
1Q00 37.9 28.3 37.3 0.5 17.2 121.2
2Q00F 40.2 20.2 32.7 0.6 6.3 100.1
3Q00F 33.8 19.7 37.9 0.3 10.8 102.5
4Q00F 31.7 19.1 40.2 0.6 5.1 96.8
Total 2000F 143.5 87.5 147.9 2.2 39.5 420.6
F: Salomon Smith Barney forecast.
Source: Salomon Smith Barney Euro Supply Monitor, 24 March 2000.

17
 In the first quarter, bond issuance was down by 21% and we expect a further decline of
12% in the second quarter. However, for the year as whole, we are looking for
only a modest fall in issuance as some States shift the financing pattern to the longer
maturities.
Thus the 30-year sector is expected to see issuance expand by nearly a third. For the
future, more debt managers may be inclined to experiment with such long maturities
now that there is a defined yield curve in euros. The inversion of the yield curves in
the UK, and now the US, illustrate the demand for long bonds from actuarially driven
investors. The flattening of the euro curve is already noticeable and may whet the
appetite of debt managers. Interestingly, just the issuance of 30-year bonds in
Euroland would amount to about a third of the entire stock of long-dated UK
government bonds where the yield pattern has become seriously distorted by
heavy pension fund demand. That has combined with low government issuance
even after the assistance from the new borrowing plans of the UK Government.

An initial assessment of the impact on monetary policy

The euro-denominated bond market has developed powerfully since inception

but this was from a low base and in a financial system that is largely based on
intermediation by banks. It is worth considering some of the salient developments
in the context of an M3 total of ELOOLRQDWHQG7KLVLQFUHDVHGE\
ELOOLRQGXULQJ

Underwritten corporate bond issuance quadrupled to about ELOOLRQ%XW

that includes a wide variety of financial corporations and there is no way of
gauging the extent to which the net expansion substitutes for bank credit.
The securitisation of bank balance sheets- via CLOs seems an important
trend but was only 0.5% of aggregate balance sheets so far. So it was not a big
impact.
The biggest sector of the new bond market is Jumbo Pfandbriefe where the
collateral is public sector loans. The real monetary impact would come
through any expansion of credit to the public sector but this is constrained by
the Excessive Deficit Rule/Stability Pact so shifts in the intermediation
channels should have only modest effects. Maturity changes in the underlying
public sector debt would be more important but there is no information on
this.
Without making extreme assumptions, it is not clear that these factors could
yet sum to an amount that would have a significant impact on broad money.
Crucially, monetary policy is not mechanistically determined by
monetary growth, so the risk of a significant distortion of policy seems
modest at least in the brief period of the euro bond markets existence so
far.

18
 PART III: The potential impact of the future markets on the operation
of monetary policy.

Even if the conclusion is that the euro bond market has probably not had much of an
impact on the conduct of monetary policy in 1999, it could be wrong to conclude that its
evolution should not be carefully monitored in the years ahead. Several powerful drivers
have yet to have anything like their full impact on the development of the bond market:
In a stable-price world, many aging investors are likely to prefer the stability
of fixed income returns for much of their portfolio. With a positive yield
curve, they are likely to be attracted to longer maturity assets to match rising
longevity.
As nominal interest rates seem low by recent historical standards, many
investors will seek to maximise their final return by dis-intermediating as
many transaction costs as possible.
Bonds may come to be seen as closer to deposit substitutes as these costs, as
well as depositary costs, fall so that it is cheap, quick and simple to convert
bonds into money.

These driving forces mesh neatly with public policy priorities. One of the strategic
ambitions for the EU, in the world of the euro, should be to make cheap and
convenient bond finance readily available throughout Euroland which naturally
involves issuing bonds across borders. The future market is likely to include selling
bonds to individuals though there are difficulties, such as different prospectus
requirements. Also, technology is bringing Internet trading of these securities into the
picture just as EMU gets underway. This co-incidence is likely to have a profound
influence on the way these markets develop.

With government issuance constrained by the drive for balanced budgets, the arrival of
the euro should open up new vistas that could take the euro-denominated non-
government bond market, over the years, up to a comparable size with its US domestic
counterpart if securitisation develops to the same extent. Using the Salomon Smith
Barney Broad Investment Grade Indexes as a proxy, the value could rise from the current
ELOOLRQRIERQGVGHQRPLQDWHGLQHXURDQGDFFHVVLEOHWRPDMRULQYHVWRUVWRQHDUHU

Such a development would still only convert less than 20% of the total
ELOOLRQ

assets of EU banks into securities.

19
Figure 8. Bond Markets Euro versus US (Euros in Billions) (Minimum Issue Size 0LOOLRQ

5000
For Govt/Agency/Gvt Gtd
4500

4000 Utility & Industrial

3500
Finance& Asset Backed
Credit Sector
3000
Pfandbrief
2500

2000 Mortgage

1500
Supranational
1000 Govt Sector
Domestic Agency
500

0 Domestic Govt
Euro Mar 2000 U.S. Mar 2000
Source: Salomon Smith Barney Fixed Income Indices, April 2000.

The advent of the euro opens up the possibility of consolidating Europes capital markets
into a single, world scale market that can intermediate the flow of savings to a wide
spectrum of end-users. The financial services industry is that intermediary so it faces a
huge challenge - and opportunity - from this same combination of rising savings flows
and new technology. The pace of consolidation in the European industry suggests that the
process of change is well underway. However, a merger takes time to be implemented
so there may be a lag before the new services are rolled out and marketed to the ultimate
client.

What is the financial services industry?

Essentially, it is those activities that give practical meaning to two of the economist's
definitions of money: a means of payment and a store of value. The first function
the payments system is probably about to undergo a revolutionary change anyway
as electronic money takes over many payments activities. The second function is the
"savings business". Once money has been collected into an account with an
institution, it must be put to work to earn a return for the saver. The route may be
the traditional bank making a loan, or a purchase of shares via a unit, or
investment, trust or by purchasing bonds. The intermediary for these actions can be
a bank, unit trust, pension fund or life insurance company.

The challenge is to maximise the efficiency of the intermediation process by utilising the
opportunities created by dramatic technological change. This could become a significant
competitive advantage for the EU within the global market place for trading securities.

20
The key areas are the steps in the process of fulfilling a transaction:
1. Formation of a legally binding contract to execute the transaction
2. Conduct of business rules
3. The actual trade - on or off a Stock Exchange
4. Delivery of the securities to/from a depository
5. The payment

The eventual aim of electronic trading is to minimise costs by making these steps
"straight-through" from investor, to broker, to exchange, to custodian/paying bank and,
finally, the security depository. This reduction in costs should widen access to capital
markets for issuers and investors.

The financial services industry is seeking to position itself as this intermediary by

offering asset management services acting as an institutional investor- or as a direct
execution agent. But the execution infrastructure is undergoing profound change as
traditional Stock Exchanges face competition from the Over-The-Counter (OTC) traders
and also from Alternative Trading Systems (ATS). Increasingly, these are Electronic
Crossing Networks (ECNs).

A note of caution: The potential removal of other Stock Exchange functions would leave
the certainty of settlement as the remaining key activity. But the scale of settlement risk is
now so great that shorter settlement periods (eventually moving to real time) are reducing
even that function. The growth of clearing houses e.g. for the futures exchanges,
illustrates one approach to change. As more exchanges become electronic and remote
trading becomes possible, it is essential for systemic stability of the financial markets for
trading functions to continue to be carried out by adequately capitalised and regulated
exchanges, clearing houses and financial intermediaries. Permitting unregulated and
thinly capitalised software or technology companies to provide key processes in
electronic securities transactions (e.g., trading software or electronic linkages) creates
new risks.

21
Figure 9. The New System

INDIVIDUALS
Order Flow

INSTITUTIONAL
INVESTORS

BROKER

Execution S T OCK ATS

E XCHANGE OT C

PAYMENTS +
Settlement

SECURITIES DELIVERY

CUSTODIANS

DE POS IT AR IE S

However, the backdrop to this change is the political drive to liberalise the capital
markets - with a 2005 deadline, set at the Lisbon Summit this March. To re-state the
conclusion of Part I of this paper, The revolutionary changes that are underway in the
market infrastructure should have come to fruition on a similar time scale, so the
scene is set for a transformation of the European financial services industry.

Given the power of the forces that are driving investors, these changes will also affect the
conduct of monetary policy. The impact will certainly come at an operational level. But
more importantly, the political setting may change radically:
Lengthening bond maturities means that monetary policy actions should not
be constrained by the risk that rising short-term interest rates would
precipitate a breach of the 3% deficit/GDP limit by EU governments. So that
loosens one constraint on policy.
But the impact on the easily assessed, market value of electors portfolios may
create additional sensitivity.
However, such a large market will eventually make it easy for savers to
hold both the ECB and politicians to account on the outlook for both
inflation and public finance via, effectively, a rolling referendum.

22