How should international financial service companies be regulated?
22 September 2004
Speech by Callum McCarthy
- For me, the timing of this lecture could not be more
propitious, as it falls on the first anniversary of my taking
up my responsibilities as chairman of the FSA. I have learnt
much in the last 12 months, and have been involved in many
issues which are both important and of wide public interest:
the challenges of establishing consumer competence and confidence
so that, with correct information and proper behaviour by
suppliers of financial services, there can be developed
a market for retail financial services based on the principles
of an efficient market rather than intrusive regulation;
the redefinition of capital adequacy requirements for banks
now being implemented under the banner of Basel II; the
fundamental reform of the regulation of life companies so
that regulatory requirements become more realistic, and
encourage a sensible and realistic matching of liabilities
and assets rather than being a somewhat arbitrary regulatory
hurdle which did not necessarily encourage sensible and
prudent behaviour; the policy on enforcement which underlies
a regulatory regime. All these – and many other issues
– are important, current and worthy of public discussion.
- In selecting a topic for this evening's lecture, I wanted
to choose one which also related to the life of Sir Thomas
Gresham. That life had many episodes which provide links
with subjects close to the concern of the FSA today. There
is Gresham's Law itself – like many eponymous concepts,
not originally formulated by the man whose name it bears:
a mark of the fact that publicity rather than original thought
even in the 16th century could prove important. Less well
know is Gresham's activities on behalf of the Crown to support
the exchange rate in Antwerp, where by 1552 the interest
on the Crown's debts was £40,000 and the rate of exchange
had fallen to 16 Flemish shillings to the £1 sterling.
Gresham dealt with this challenge in various ways, but principally
by spending £1,300 per week of Exchequer money in
the Antwerp market to prop up the value of sterling, through
secret money market operations. He bought, on a daily basis,
small sums of sterling in undisclosed transactions –
as he described it "it shall not be perceived nor shall
it be no occasion to make the Exchange fall." Gresham
subsequently had the task of repatriating – in the
face of a prohibition on bullion exporting – his gold,
a task which he solved by the practical device of smuggling
it out to England packed (and concealed) in armour.
- Faced with this history, I was tempted to use the link
with Sir Thomas Gresham to discuss either the issues surrounding
market manipulation or those surrounding money laundering.
I decided however that to concentrate on these less savoury
aspects of his career would be unattractive. What I want
to discuss is the set of broader questions associated with
Sir Thomas Gresham's international activities, namely in
a world where international financial services companies
are increasingly the norm, and where cross border transactions
ever more important, how should regulation develop to match
this increasing internationalisation? The reality of the
question is clear. I am speaking this evening in Canary
Wharf, where we have the offices of such international financial
services firms as Morgan Stanley, Lehman Brothers, Bank
of New York, Bear Stearns, CSFB and Barclays. There are
many more firms in the City of London, including Deutsche
Bank – which employs more in London than it does in
its headquarters in Frankfurt – and Standard Chartered;
some of which are the corporate headquarters (although their
business may be more outside the UK rather than here), some
are branches of businesses headquartered elsewhere, some
are UK incorporated subsidiaries themselves with branches
elsewhere in the EU. To illustrate this further, in one
of the towers behind me here at Canary Wharf, there are
the global headquarters of HSBC which has operations in
some 80 countries and territories around the globe. Over
two thirds of the HSBC group's profits come from outside
of Europe, well over two thirds come from outside the UK.
- In the other high tower behind me is Citibank International
Plc, Citigroup's UK subsidiary and the headquarters of Citigroup's
European activities. Citigroup, like a number of other US
firms, makes extensive use of the ability to 'passport'
throughout the European Union. It has established branches
in Austria, Belgium, Denmark, Finland, France, Greece, Ireland,
Italy, Luxembourg, Netherlands, Norway, Portugal, Spain
and Sweden. Running these international businesses poses
difficult and important questions of management and control
for their management teams. They also raise difficult and
important questions for regulatory authorities. It is these
questions I wish to address this evening.
- The issues which arise are common across financial services.
They are not confined to banking, but arise equally –
albeit in different forms – in both insurance and
securities trading as well. They include both prudential
issues – how to ensure that companies and groups,
and the component parts of groups, are properly controlled
and supported by adequate financial resources - and conduct
of business issues – how to ensure that products and
services originated in one country and distributed in others
meet the standards required in all those countries. These
are issues which are being discussed in particular in the
context of the EU, but they arise equally in the wider global
context: the supervision of Deutsche Bank, of Citigroup,
or of HSBC raises issues which clearly require answers not
merely within Europe, but worldwide.
- Before I start this discussion of the issues, I should
comment on one complicating factor. In financial services,
as in other walks of economic activity, it is not unknown
for regulatory issues to be used as camouflage for national
protectionism. I find it hard to eradicate the suspicion,
for example, that some countries' attitude towards the real
supervisory issues associated with the relocation of supporting
services to cheaper locations – either within the
group or by outsourcing – is informed by a wish to
preserve, at least in the short term, jobs at home. I intend
to ignore these questions of base nature. And I would claim
that in so doing I am reflecting in discussion what has
been the reality of British regulatory policy for financial
services. The FSA is concerned with efficient, orderly and
clean financial markets, not with protectionism. We are
indifferent to the nationality of the owners and the managers
of the institutions we regulate.
- Let me start by describing the present arrangements for
supervision of financial services companies across borders.
These arrangements are built on twin pillars, one personal
or sociological; the second administrative. The former is
what I sometimes describe as the fraternity of central bankers
and regulators. By this I mean the mutual trust, shared
vision, confidence and consciousness of mutual dependence
– as well as of conflicts of interest – which
exists among central bankers and, where they are different,
financial supervisors. They share an understanding of the
issues; recognise the special nature of the political territories
– different in every country – in which they
operate; are conscious of the constant need to balance independence
and accountability; and are aware of the scale of their
responsibilities, both prudential and conduct of business.
These bonds are reinforced by the repeated nature of the
contacts between us: at the BIS in Basel; in repeated European
meetings; in more reflective circumstances such as the annual
Jackson Hole extended seminar; at the FSF in whichever country
it is meeting. The constant travel which is an inescapable
part of the life of any of us who belong to this fraternity
has a purpose, and produces a result. It renews and strengthens
relationships which matter, and which would otherwise weaken
without trust. I am very conscious that relationships which
have been important in the past are subject to generational
rather than regime change. Institutional relationships can
and do endure even though individuals move on. It is the
responsibility of the individuals currently in place to
nurture and develop these long term links. In London, one
of the world's most important capital markets, in the last
2 years we have seen the arrival of a new Governor and two
new Deputy Governors. Both John Tiner and I are as new in
our present responsibilities. We are about to lose Michael
Foot. At the New York Federal Reserve Bank, Tim Geitner
is as new in his job as President, and Gerry Hawke is soon
to retire as Comptroller of the Currency. Renewing and strengthening
the links with other senior financial regulators is an important
part of the discharge of my duties.
- The second pillar of the prudential regulation of international
financial services is the set of administrative arrangements
which fall under the general title of home and host regulator.
Those of us involved in regulation are aware of growing
calls from the financial industry for a streamlining of
regulatory arrangements in the form of an extension of the
so called 'lead supervision' concept. The broad principle
of lead supervision is simple and not particularly new:
for a group the principal responsibility should lie with
the home regulator in the country in which the group is
based, with a subordinate set of responsibilities attached
to the supervisor in the various host countries in which
the group operates. On this basis, the principal regulatory
responsibility for Deutsche Bank lies with the German regulator
BaFin, the principal responsibility for HSBC with the FSA,
and the principal responsibility for Citigroup with the
Federal Reserve Bank of New York. The appeal of this approach
is clear: in prudential policy, there is the prospect of
one regulator taking a view on group wide capital adequacy,
controls and the overall management of the group, with no
need to duplicate this country by country; in conduct of
business policy, there is the prospect, for those countries
which recognise each others' standards, of one test of consumer
protection being applied and the product then being accepted
by other countries – "passported in" as
the jargon has it. It is not surprising that many international
financial services companies have enthusiastically embraced
the idea that regulation should be streamlined, whether
this involves a single home regulator, a lead regulator
or a consolidated supervisor.
- The appeal of some of the more extreme versions of lead
regulation is obvious: this would involve replacing, all
or in part, the present multiple regulatory relationships
which a major group has with a single relationship. On the
face of it, this would be deeply attractive. But the reality
is more complex. This evening I want to dissect some of
the realities, not in a destructive mode but in the belief
that any attractive concept, unless tested and proved against
reality, remains merely an attractive concept. If we are
to go beyond simple slogans of support for primacy of the
home regulator, or the need for consolidated supervision,
we need to test what is involved.
- Let me start by considering what is needed if a host
state regulator with (actual or perceived) responsibility
for the local activities of a global institution is to have
real confidence that a home country regulator will be able
to discharge its lead regulation role effectively. There
are at least four questions which have to be addressed in
any discussion of the extent to which one regulator can
be entrusted with the discharge of duties on behalf of another
country's regulator. They are:
- does the lead regulator have the legal powers needed?
- is the lead regulator willing to do the task?
- is the lead regulator competent to do the task?
- if something goes wrong, can the problems get dealt
with quickly and fairly by the combination of regulators
involved.
Let me deal with each in turn.
- It is only possible for a home regulator to discharge
his responsibilities if the regulator has the necessary
legal powers. That is a truism. But the fact which is often
ignored is that there is no congruence between the powers
or structures of different regulators. There is no agreement
on the regulatory model across the world, along two dimensions
of difference. First, there is disagreement as to whether
a single organisation should have both prudential and conduct
of business responsibilities – what I would call the
north:south divide. Some institutions – the Dutch
securities regulator, for example – have conduct of
business responsibilities only; others – the French
securities regulator, for example – are responsible
for both prudential and supervisory issues. Second, there
is no agreement as to which sectors should be covered by
a particular national regulator – what I would call
the east:west divide. Some regulators – the FSA, BaFin,
the Japanese FSA are among the growing number in this camp
– regulate insurance, banking and securities; some
– the Dutch central bank, APRA in Australia and OSFI
in Canada where securities remains a provincial-level responsibility
– regulate two from these; other countries –
China, France and the US are examples – have separate
regulators for each. With these two dimensions of difference
in legal powers between regulatory organisations, it simply
is impossible to establish in all cases home regulatory
responsibilities for all the functions of an organisation.
Too often, the home organisation does not have the powers.
- There is also the question of policy: the willingness
of a home regulator to discharge the responsibilities which
the host regulator would like to see. There are occasions
when standards differ between nations: it is clear, for
example, that the requirements for issuing a prospectus
differ even within Europe and that this reflects not only
legal differences but also, and more importantly, national
decisions on the balance of advantage to a country of tighter,
or looser, consumer protection.
- Third, there is a simple question of competence. I am
fortunate, in that my responsibilities are met by a staff
of 2300 people and a budget of £200 million. The task
of regulating well – at a minimum, for a European
regulator of responding to the range of responsibilities
imposed by EU financial services legislation – is
a demanding one. I would find it very difficult to discharge
this responsibility easily or competently with the small
staff and limited resources of many financial services regulators.
I do not doubt their individual quality, nor their willingness,
but in some cases the resources are simply not up to the
task of acting as the home regulator for a large group.
There are many issues, some now before the British courts,
raised by the events, now some decades past, of the supervision
of BCCI. I want to comment on a specific aspect, namely
that the experience of regulating an international bank
with many different regulatory problems from the resources
then available in Luxembourg shows that the simple assumption
of the practicality of relying on the home supervisor is
flawed.
- The final question is what happens if and when something
goes seriously wrong with the international group. In particular,
can the various regulators and others sort out the problems
quickly, effectively and fairly? Fortunately, we have not
seen this tested often yet, which is just as well because
it is an area of great potential concern. What we did see
in the liquidation of BCCI was a group go into liquidation
in nearly 40 jurisdictions and a wildly different treatment
of local and overseas creditors in the process. British
bankruptcy law is very even handed as between UK and other
creditors. Much bankruptcy law elsewhere is not. And a host
country regulator cannot but be aware of the likely fall-out
from this fact. Similarly, when Barings went into administration,
a spotlight was briefly thrown onto the complexities of
different trust laws across the various jurisdictions in
which Barings operated. ING's purchase of the group fortunately
put the spotlight out before things got too difficult. But
no one could be confident that the failure of a major international
group now would not cause market and/or settlement log-jams
across the world and provide lawyers with decades of work!
- It is therefore important to recognise the limitations
on the simple reliance on home and host regulator as one
of the twin pillars of international financial regulation.
First, the home regulator may not have the necessary legal
powers. Second, the standards applied by the home regulator
may, as an act of deliberate policy, differ from those in
the host country. Third, the resources available to the
home regulator may be inadequate for the task. For all those
reasons, we need to think carefully before accepting simplistic
models involving complete reliance on home supervisors as
a practical solution to the legitimate needs of industry
for efficient and non-duplicative regulation.
- There is a further reason for thinking carefully about
this principle. We are increasingly facing situations where
financial services are of particular importance –
sometimes arguably of systemic importance – in a host
country, but are of less importance in the home country.
Within the EU, for example, foreign institutions account
for more than 70 per cent of the bank systems of Slovakia,
Estonia, Poland and Hungary; the fund management sector
in Hungary is overwhelmingly foreign owned; in the mid 1990s,
the failure of the Meridien Banking Group (which was registered
in the Bahamas) led to widespread chaos across Africa, in
countries like Zambia where the group controlled a large
domestic bank. These examples are relatively far from home
but the logic of the single market and the legislation underpinning
it – such as the European Company Statute –
make it likely that this unconventional pattern of relative
importance in home and host markets will become increasingly
common. You may ask why this should be a concern. In these
cases, there is simply the problem that the host regulator,
if relying on the home supervisor, simply would not control
the great majority of financial activities in his own country
– the task he is legally charged and politically expected
to discharge – an awkward position. But the position
can be even more acute. Consider what would occur if there
were a problem affecting Nordea, a bank headquartered in
and regulated from Sweden, but arguably of greater importance
to financial stability in Estonia and in Latvia than in
Sweden. Were there to be a group wide problem, the Swedish
regulator, the Swedish central bank and the Swedish finance
minister would be called upon to make the decision of whether
to mount a rescue and, were they to decide to do so in a
way which went beyond a market based solution, ultimately
the Swedish taxpayer would meet the costs of it. It is easy
to see that this misalignment between the interests of those
paying for a rescue and those benefiting from it could be
a source of acute difficulty. Now I hasten to stress that
I use Nordea as an example simply to illustrate the general
problem. I am not implying in any way that the scenario
I have sketched is in any way likely – but it, or
some other example, is possible. And were this, the "Nordea-type"
issue, to arise the adoption of an over-simplistic lead
supervision model could well be found to complicate things
rather than simplify them.
- I should mention one other situation which not only might
arise but, in my judgment, is more likely to arise: that
is the question of the governance and the supervision of
exchanges and of clearing, settlement and payment systems
operating on a cross border basis. I suspect that, just
as regional exchanges in Britain have disappeared, the same
economic pressures will result in consolidation within the
present number of European exchanges, with the resulting
issue of how they are to be regulated. For example, were,
as is often rumoured as a possible transaction, Deutsche
Börse to acquire the London Stock Exchange (a possible
transaction on which, I should make clear, I am expressing
no overall judgment whatsoever) there could be the possibility
that (if the LSE were closed and all UK services provided
through Deutsche Börse screens) the home regulator
would be the Land of Hessen (BaFin has no direct regulatory
responsibility for Deutsche Börse). It is unclear that
this arrangement would prove entirely adequate. And it is
interesting that the most significant examples of cross
border market infrastructure providers, the LCH Clearnet
and Euroclear arrangements, have been dealt with by multilateral
agreement between the member states regulators involved,
with rights enshrined for all, rather than by reference
to home and host responsibilities.
- My purpose until now has been to expose the difficulties
in adopting a simplistic approach to home and host supervisory
issues. Those who argue for the complete supremacy of the
home supervisor need to recognise the complexities and realities
with which any set of supervisory arrangements must cope.
- It is relatively easy to show that simplistic solutions
do not match the complexities and realities of international
financial services. But if not the simplistic answer, what
is the more useful approach? For I recognise that the problems
are real: how do we help HSBC deal with more than 150 regulators
in 80 countries? How do we encourage pressure, within the
EU, for a single market which comes from the principle of
freedom of establishment? How, globally, do we prevent a
balkanisation of regulation, with capital requirements being
imposed on a national basis in a manner which proves economically
inefficient? These are real questions, even if the simple
answer of home regulation is inadequate.
- My own answers to these questions are based on three
approaches. The first is obvious, and is the point that
I have made at some length in this lecture: we need an approach
which is realistic, not one based on a simplistic search
for a single regulator for each financial services group.
I should add that I am ignoring the many issues arising
from corporate structure in general, and the use of subsidiaries
versus branches in particular. This is worthy of a lecture
in itself – and indeed recently was. I commend for
any interested in this important topic the recent lecture
by my Belgian counterpart, Eddy Wymeersch, "New Perspectives
in Financial Supervision in Europe".
- My second approach is to recognise that not all financial
institutions are the same, and therefore that not all supervisory
issues are the same. The position of a bank with a minor
presence in a host country differs from that where the presence
is major; and differs again from that where the bank is
large enough to pose systemic risks within the host country.
We need to recognise this. Similarly, we need to recognise
that the issues posed by a bank will differ from those posed
by an investment manager, which in turn will differ from
those posed by an exchange. There is already some welcome
recognition of these realities: the Commission has recognised
that the home:host model might need to be modified for the
"specificities" of clearing and settlement; and
the actual supervision of Euronext and of Euroclear is based
on a college of supervisors, not on the home:host principle.
At present, these departures within Europe from the strict
home:host principle have occurred on an ad hoc basis (the
Euronext decision, for example, was a Dutch Ministerial
decision). Going forward, we need to recognise the variety
of institutions and situations we face, and develop a taxonomy
for measuring and judging the appropriateness of the home:host
model – or, to put it more simply, to decide how regulators
can best co operate across national borders. In some cases
it may well be appropriate to place extensive, if not exclusive,
reliance on home supervisors. In other cases host supervisors
– for example of high impact branches – will
need to have the opportunity to engage in more of a dialogue
with home supervisors about things that really matter from
the point of view of their markets and objectives. And in
the supervision of groups I believe that our objective should
not be to achieve wholesale changes in formal responsibilities
along the lines of the simplistic models referred to earlier.
Rather, we need to devise truly effective models of lead
supervision within existing accountabilities that involve
an appropriate division of labour in the pursuit of tailored
and co-ordinated regulatory programmes. I believe that this
can be achieved in a way that meets industry's legitimate
desire for more streamlined regulation. And I might add
that, at a domestic level, we have already gone a considerable
way down this road since the creation of the FSA in our
approach to the regulation of the largest groups in the
UK. Developing this taxonomy and devising appropriate regulatory
structures will be a major task for regulators. It will
need to be done in a European context, and will therefore
require work in all three of the Lamfalussy committees (it
is the subject of current work within the securities committee,
CESR). It will require work beyond Europe, where in banking
the obvious forum is the Basel Accord Implementation Group.
- My third approach is again based on a wish to depart
from simplistic slogans. We need to develop protocols which
define both the duties and the rights of both home and host
regulator: what information can the home regulator expect
to receive from a host colleague in respect of a financial
institution, and what information is the home regulator
required to provide to a host colleague about a financial
institution? What action can a host regulator take if the
actions of the home regulator do not meet the requirements
of the host country? What can be done to deal with home
regulators who do not, for reasons of law, policy or competence,
discharge their duties properly?
- There has already been work, formal and informal, to
establish such protocols. The formal work is to be found
in European Directives: the Financial Groups Directive,
the Market in Financial Instruments Directive and the third
Capital Adequacy Requirements all contain provisions setting
out obligations and rights for home and host supervisors
– albeit without clarity as to how "co-operation"
is to occur, or be made to happen. And there is little in
these Directives of the delineation by type of financial
activity for which I have argued. Still, it is a start on
which we can build. A different formal approach to defining
what is required may be found in the various Memoranda of
Understanding which have been concluded between different
regulators: the FSA, for example, has over 150 MoUs with
our counterparties across the world.
- Alongside these formal approaches are informal, practical
arrangements – making the system work in practice.
For UBS and Credit Suisse, for example, there are, every
six months, meetings of the Swiss Banking Commission, the
Federal Reserve Bank of New York and the FSA – the
main relevant regulators in the main three capital market
centres in which the firms operate – to compare issues,
exchange information and co-ordinate action. For HSBC, we
have started regularly to convene what we call the college
of HSBC regulators for exchange of information on HSBC's
plans to implement Basel II. These are not held on fixed
timetable but we would expect to hold meetings at least
once a year. Neither do the colleges involve the more than
150 regulators with which Sir John Bond is properly conscious
of having to deal – that would be too unwieldy. But
all the main regulators of HSBC round the world, accounting
for some 80% of its assets, are included; the college which
we hosted here at the FSA in April involved regulators from
Hong Kong, the US, France, Switzerland and Canada. We will
continue to develop such approaches, both for HSBC and other
groups for whom it would be beneficial.
- This evening I have set out some – by no means
all – of the realities and complexities, legal, political
and economic, which arise in relation to the regulation
of international financial services firms. I have argued
that any approach to devising better regulation has to take
these realities into account, and that a simple slogan of
advocating the primacy of home regulator or lead supervisor
fails to do this. I have instead argued for an approach
based on two foundation stones: first a better taxonomy
of regulation – an understanding of the issues raised
by different types of financial institutions and financial
activity; and second on developing specific protocols which
define the rights and duties which each of home and host
regulator should have, for the various activities and institutions
identified in the taxonomy.
- If we advance in this way, I believe we can make a significant
contribution to facilitating international financial activities
– something which will generally benefit London as
the most international of the world's capital markets. It
is something of which Sir Thomas Gresham, whose activities
breached sixteenth century regulations, might well not have
approved. Despite this, we need to make this advance.