Institute of International Bankers - Washington
Callum McCarthy
Chairman, The Financial Services Authority
26 September 2005
As I remember this occasion a year ago, I then argued for developing a protocol of what home regulators should expect of host regulators, and what host regulators should expect of home regulators. I want to come back to the practicalities of these issues, in terms of what information regulators should look to another regulator to provide – and, equally important, what information they should neither request nor expect.
This is obviously important to the regulatory organisations – which is the reason why so much work has gone into the practicalities of the Financial Conglomerates Directive, and of Basel II, which have been much discussed between EU and US regulators. But it is also important for the banks and investment banks that we regulate, because without efficient information sharing between regulators there is the prospect of these firms being faced with multiple but frequently overlapping information demands from the various regulators with whom they have to deal. More positively, if the various regulators succeed in co ordinating their information requests, the burden on banks, investment banks and securities firms should be lessened. So this matters for you, as well as for us.
There are, of course, formal arrangements for information exchange, typically set out in a Memorandum of Understanding between two regulatory organisations. They sometimes fulfil a legal purpose, in that they meet a condition which requires a formal agreement between regulators before confidential information can be exchanged. And there are certainly plenty of them: the FSA has already concluded around 150 MoUs with other regulators, and more are in the pipeline. But they are not, and I think will never be, the fundamental instrument on which we rely. Instead, we would always prefer to place reliance on arrangements that are less formal and much more important, namely the willingness and ability of regulators to speak frankly and in a timely manner to each other about issues that matter – what I sometimes loosely call the freemasonry of central bankers, and what should more properly be regarded as the fraternity of regulators; what in practice is the ability of regulators to pick up the phone to each other, and have a grown up conversation with a counterparty whom they know and respect.
But we need to go beyond this to establish some ground rules for how regulators use both their formal agreements and the goodwill that exists between them: what information should we ask for, how often, and in what form? And what information should we be prepared to give? How will this be affected by the particular circumstances (it is obvious that the normal day to day business as usual exchange of information will be different from that needed when there is a crisis – either institution specific, as with LTCM, or more wide, as with the events of 9/11)? How do we deal with the special circumstances of enforcement actions?
It is obvious that there is no one size fits all rule. Here, as elsewhere with supervision and regulatory questions, there cannot be a "any colour so long as it is black" approach. But there are, I think, some principles which can usefully be set out. Here are four, all of which deal mainly with information flows from home regulators to hosts:
- we should do all we can to avoid seeking what I will call "primary data" from other regulators. For example, we at the FSA should not, except in the very rarest circumstances, seek to see the inspection reports of the New York Fed, or the OCC, or the SEC of a US institution they supervise. Any regulator seeking such information would need to make the case through very formal channels. Instead, in the great majority of cases we would rely on the considered conclusions of the supervisory regulator. There will be occasions when more specific information is required – particularly in relation to enforcement actions, when details of share dealings are often central to a case – but then the information will be case specific;
- the test of whether information is requested should be set quite high: the information requested should be asked for only if it is directly and significantly relevant to the responsibilities of the host regulator. The information volunteered by the home regulator in the normal course of business should also be subject to quite a high test: it should be the essential information about a group – material factors affecting its financial soundness, which clearly encompasses both capital and also control and systems issues;
- the test of relevance must be market specific. By that I mean that home regulators need to recognise the importance of a group in – and to – countries even though that country may be of minor significance to the group overall. There is often an asymmetry: the Polish activities of Citigroup, for example, may properly not feature as a critical issue in the Fed's or OCC's oversight of Citigroup, but Citigroup's activities in Poland, where Bank Handlowy is very significant, may properly be an important concern for the Polish authorities. These regulators, like the FSA, which are constructing colleges of regulators for the most important institutions for which we have home regulator responsibility need to be conscious of this; and
- finally, we need to avoid complicating things more than is necessary. In particular, it is a good principle that bilateral issues should be dealt with bilaterally. I am aware that the Financial Conglomerates Directive makes provision for a single EU co ordinator – and I am particularly conscious that for many of the most important US financial firms subject to this directive that co ordinator is the FSA. I hope that we will be involved only where we add value: ie where there is an issue affecting several EU regulators where co ordination is required; and that when there is a bilateral issue involving the US and a single Member State it will be dealt with directly between the relevant regulators, not routed via the FSA.
You may regard these principles as so obvious as not to need saying. But they have not hitherto been said, and I think merit putting on record. In regulation, as in many walks of life, there is a lot to be said for keeping it simple.