Regulation in different industries
Speech by Callum McCarthy, Chairman, FSA
Cass Business School
13 February 2008
To have been asked to give this, the inaugural lecture in the series named after and honouring David Currie, is both a great pleasure and a considerable responsibility. The pleasure arises from the opportunity, which occurs far too rarely in life, to express appreciation and gratitude. I have had the enjoyment of working directly with David when he was a member of the Board of Ofgem, and more indirectly with him in his present role as a chairman of Ofcom, where we compare and contrast the issues with which we and other regulators grapple – of which more later.
When I was establishing the Ofgem Board, David was one of the first people I approached to ask to serve on that Board. I was keen to have the input of an economist, since one of the challenges for a regulatory organisation is to ensure that its individual decisions are taken against a proper appreciation of the systemic effects of those decisions. The right economist – one who both understands the particular issues of an individual case but also puts that decision into the context of the wider implications of the decision – contributes substantially to the quality of decision-making of the regulatory organisation. David emphatically made that contribution, and helped establish the intellectual rigour and the standing of Ofgem – a contribution by economists which I should add was continued at Ofgem by Len Waverman, and which is ably discharged on the FSA Board by David Miles. It is a great pleasure today to have the public opportunity to express my gratitude to David for the help he has given me.
What I want to address this evening reflects David's – and my own – interest and involvement in regulatory issues. I want to examine some of the issues which lie within the very wide range of activities which go under the banner of regulation; to establish some major differences between different regulatory responsibilities; and also to tease out some of the issues which are common to all regulatory organisations. I shall do this not particularly in an economic context, but rather to analyse the general characteristics of regulatory activities, and some of the implications of those characteristics.
I should start by recognising what a wide and diverse range of organisations fall within the general category of "regulators". If anyone has doubts about this, I would simply refer them to the 2005 Hampton Review of national regulators, which identified 62 national regulators which it regarded as falling within its scope, a further 16 which fell without its scope and – perhaps most worrying of all – no less than four new national regulators which had been created or proposed while the review was in process. The common feature – perhaps the only common feature – of the Audit Commission, the National Bee Unit of the Central Science Laboratory (an organisation from which, as a beekeeper, I receive infrequent inspection: it is, I think, my only current direct experience of regulation as seen by the person regulated), the Charity Commission and English Heritage is that they are all, in their very different ways, national regulators – as are both Ofcom, chaired by David, and the FSA, which I chair. Diversity is certainly the name of the regulatory game.
It is worth dwelling a moment on the categories into which such different organisations fall, and the motives for establishing separate regulatory organisations rather than their duties being carried out by central government – as many regulatory activities are in fact carried out by central government. In some instances, I think we should recognise that the regulatory organisation is best regarded as an agency of central government, carrying out activities which can be undertaken largely at the behest of government without a clear need for ‘independence’ from Government. I would certainly put the Environment Agency in this category, and tend to place the Health and Safety Commission in the same category (although its makeup is an affirmation of the shared belief of employers and employees of the value of safe practices). There may well be benefits of concentrating expertise within a special agency, or freedom from (or at least relaxation of) constraints on public sector pay policy, which make this an efficient arrangement. But I confess that I find it hard to identify the logic, as distinct from the historical explanation, that takes one large part of Defra responsibilities – environmental – and places them in an agency, and takes another large Defra responsibility – the detailed and often intrusive regulation of farming – and discharges a great deal of this set of responsibilities within central government. There are good historical reasons – but they should not obscure the fact that some regulators are much closer to government than others.
I have described the position of the Environment Agency in terms of closeness to government. But – as the contrast between the Environment Agency and Ofgem makes clear - a more common feature of regulatory organisations is the deliberate separateness from government which they enjoy – though, as I will show, enjoyment is not always the dominant emotion. In some instances – as with the economic regulators established when state-owned monopolies were privatised – the separation had a very direct economic imperative. I do not think it would have proved possible to attract investment in these industries if it had been left to the government to set limits on their prices after privatisation; and equally it was clearly necessary to set limits on the prices charged for services provided by what had been state monopolies – hence the creation of Oftel, Ofgas, Offer, the Water and Rail Regulators.
In other instances, the motivation for the separation of regulatory organisation from government is more diffuse than the specific need for the government to avoid setting prices for an undertaking which it is selling. There is instead a desire by government to set general policy but to leave its detailed implementation – the practical task of regulation, what, in financial services, was realistically described by the FT as "the grubby business of regulation" – to independent regulatory organisations. It recognises the difficulty, if day-to-day regulatory decisions are made within central government, to avoid direct ministerial responsibility, even though efficiency requires those decisions to be delegated. Allocating those to an independent regulatory organisation allows this separation – something which I believe is relevant to the raison d'etre of many regulatory organisations. I think that this is undoubtedly one explanation of why difficult decisions on authorisations, capital adequacy, the appropriate stress tests for banks and the like are the responsibility of the FSA rather than HM Treasury.
There is also a question of perception. An independent regulatory organisation is – correctly – thought to be less susceptible to political influence than is a government department. It is this perception which lies behind the allocation of various standards setting and standards judging responsibilities to independent regulators rather than to government departments. In their different ways, the Audit Commission, Ofsted, and the Office for National Statistics are examples of this.
And last, I would nominate some activities of some regulatory organisations as simply dealing with problems which are too charged for politicians to wish to handle: the questions, often fiercely contested on the battleground between religious belief and scientific ambition, of stem cell research which are a part of the responsibilities of the Human Fertilisation and Embryology Authority, exemplify this. In this part of its work, I regard the Human Fertilisation and Embryology Authority as acting rather like a permanent Royal Commission, seeking to bring facts and logic to questions which ultimately have to be answered by reference to value systems.
I have spent some time in deconstructing "regulation" into different parts because I believe that the requirements demanded of regulatory organisations – and also the qualities required – reflect these very different purposes and backgrounds. For example:
The possibility – indeed probability – of conflict between government and regulatory organisation is inherent in the system of independent regulation, and should be recognised and understood not as a failure of the system, but rather a necessary and indeed desirable feature, which should be managed carefully, not simply decried. But the tensions are likely to be greatest when the government still retains a direct economic interest in the regulated organisation. I do not believe the tensions between rail regulator and government which were such a feature of events of some years ago are most usefully described, as they sometimes are described, in terms of the personalities involved. Rather their explanation should be sought in the very real economic interests at stake – something which also underlies the relationship between government and Postcomm;
Second, there are great differences between those regulatory organisations which deal with structure and those which deal with behaviour. At Ofgem, apart from price control of the national monopolies, I dealt mainly with structural questions: how to separate those parts of the energy companies which comprised the highways for electrons or gas molecules from those parts of the energy companies which supplied electrons or gas molecules. At the time of privatisation, these two sets of activities were carried on by the same firm so that, for example, British Gas both owned the pipelines and supplied gas. Given its interest as a supplier of gas, it had no incentive to make it easy for competing gas suppliers to access its pipeline monopoly – indeed, quite the contrary. Attempts to compel British Gas to give open and fair access to its pipeline system to its competitors – the initial regulatory position post-privatisation – by behavioural regulation were always bound to be constrained by the fact that it was an attempt to compel an organisation to act against its economic interests. Regulation of this nature tends to be both intrusive and detailed in nature, and bruising in character. I am glad to say that by the time responsibility for gas regulation passed to me a more sensible regulatory compact had been agreed, based on structural, not behavioural, principles. British Gas separated its businesses into, on the one hand, a monopoly pipeline owner, incentivised to carry as much gas as possible with efficient use of its capital invested, irrespective of who supplied that gas, and on the other hand a quite separate business, with no interest in the pipeline system, supplying gas, which competed with other gas suppliers. Regulation then principally became concerned with the price controls for the pipelines; the whole of gas supply, now a competitive activity without advantage for the incumbent, could be left to competition, without any need for price controls. Regulatory activity could be severely scaled back. Some 70 per cent of the gas and electricity activities subject to price controls at the time of privatisation were released to be set by competition in a market that was and remains fiercely competed for. (I would add that, as well as structural reform allowing market forces rather than regulation to determine prices, it brought much greater clarity to the two firms which were created from British Gas – not surprisingly, since the managerial skills, the management focus, and the capital structure required by a monopoly pipeline owner and by a competitive gas supplier are all very different. Separation allowed specialisation, improved performance, and improved returns). Comparable structural changes occurred in the electricity industry. At the FSA, however, few such structural remedies are available, for the reason that conflicts of interest within financial services are more endemic, pervasive and incapable of being eradicated: an obvious example is that any organisation which acts as both principal and agent clearly has conflicts – and there are many, many others. In financial services, it is simply not possible to deal with most conflicts by structural reform, and instead behavioural regulation is called for – hence, for example, the rules on best execution, on firewalls between activities, on the precedence determining which orders are filled when – in short, the mass of detailed rules needed to deal with specific conflicts which cannot be eliminated through structural reform. It makes for detailed and intrusive regulation, as behavioural regulation tends to be. The work of the FSA for this (and other) reasons is more difficult than that of Ofgem;
Third, there is a very significant effect on the regulatory relationship – the relationship between regulator and regulated – which arises from the responsibilities of price regulation. Just as – and, I think, for similar reasons – the relationship between regulator and government is more charged when the government retains a substantial economic interest in the direct results of regulation, so the relationship between a regulatory organisation and firm is more charged if the firm is subject to price controls. This, which may appear an obvious point, is not trivial in its effects. Good regulatory relationships are based on a high level of openness on both sides: clarity and candour from the regulated firm, and from the regulator too. But I think it is necessary to recognise that the rather naked economic interests exposed by a system of price control inhibit this candour. There are some very public examples of this: last week Ofwat confirmed that they have fined Southern Water £20 million for "deliberately misreporting customer service performance to Ofwat and systematically manipulating information to conceal its true performance". But there are also less extreme but in a wider sense perhaps as damaging examples. Consider the events of the 1994 price review for the electricity distribution companies. This price review had to be reopened after a particular company (Northern Electric), in marshalling its defence against a hostile takeover bid, showed that it could throw off cash on a scale and at a speed quite contrary to what it – in common with all its other peers – had claimed when arguing what was possible in previous price control discussions with the electricity regulator. I confess that I find these events difficult to reconcile with a need to have honest discussions – a point I attempted to make clear to the companies concerned when I became responsible for the next price review of electricity distribution companies in 1999. Indeed, if I am allowed to reminisce for a moment, I recall warning the heads of the US-owned distribution companies (as many then were) at a dinner in Washington that I expected a more honest set of discussions than had occurred five years before. Unfortunately, to make my point I used an English colloquialism incomprehensible to Americans: my actual remark was "Last time you legged over my predecessor. I don't expect it to happen again." Luckily, alongside the US chairmen was the British chief executive of one of the companies, who helpfully explained what I meant by the expression "legged over". He then – because he was an honest man – remarked: "Callum's right. We set out to deceive Stephen [the previous electricity regulator] and after the event were embarrassed by the extent to which we succeeded". His was an honest, if indiscrete, description of what had occurred – and makes my point that price controls can easily corrode the relationship that should exist between regulator and regulatee. I should add that, quite apart from the general harm to the regulatory relationship that this produces, it is particularly harmful to the process of setting price controls. Setting a price control, like setting a budget within a firm, is a process which relies on standards of honesty on the part of those seeking resources, and fairness from those allocating resources. It is also a process which is repeated – every five years for energy price controls, annually for the budget of a firm. It is not very difficult for anyone bidding for resources to make dishonest claims – to underestimate cost savings available, to overestimate capital requirements – and to succeed in having these accepted – once. But any temporary success comes at a much greater long-term cost. The 1994 distribution price settlement had to be reopened by Offer (to the disadvantage of the companies); and companies which had proved so unreliable in assessing their future cash flows were totally incapable of arguing with any credibility against the imposition of the windfall tax imposed on them by the new Labour government in 1998. I am glad to say that, by the time of the 1999 price review, the electricity distribution companies had for the most part become reconciled to the advantages of greater candour and more realistic figures – both to their long-term advantage and to the undoubted improvement in the regulatory dialogue. But I repeat my general point: price controls add a tension to the regulatory relationship, and in extremis can effectively destroy it.
I have set out some of the particular characteristics of regulation: the special strain on the inherently stressed relationship between regulatory organisation and government when the government retains an economic interest in the regulated industry or firm; the difference between structural and behavioural regulation, and why the former is easier – for both regulator and regulated; and the particular problems, and threat to the regulatory relationship, arising from price controls. Let me now turn to a general issue for regulatory organisations, namely how to retain and develop their legitimacy.
Let me state the problem. Individual regulatory organisations are given the power to take decisions which can have great social and economic import; those decisions often confer advantage to particular industry sectors or particular firms (who understandably do not acknowledge the advantage conferred) and disadvantage others (who do complain vociferously – again understandably); these decisions are often far from straightforward, and inevitably not all are correct – certainly not in retrospect – and are on this basis understandably subject to criticism. Overshadowing all these problems is a regret among I think many politicians and officials that important decisions are no longer theirs – something which is enunciated publicly from time to time in relation to the independence of the Monetary Policy Committee, and which, when I was at Ofgem, was quite frequently clear in private. It is not surprising that officials who have had no experience of the time when they had to advise on and subsequently defend a wide range of difficult decisions, and who now have to intercede between independent regulatory organisations and Ministers whose perfectly proper ambitions have no legal mechanism, absent primary legislation, to achieve those ambitions, should be sometimes nostalgic for the previous era. So all regulatory organisations are subject to erosion of their legitimacy, as understandable and sometimes justified specific criticism reinforces what is a general problem of ensuring that the reason for granting independence is not forgotten.
I think the question of legitimacy is a central question for any thoughtful regulatory organisation. Let me set out my own conclusions on some general issues, and discuss one important element – enforcement – in more detail.
First, any regulatory organisation must cleave to the statute which established it – not only because this is what the law demands, but because this is the bedrock of the organisation's legitimacy. It is the link to the democratic process. It provides powers, duties and the basis for and limitations on independence. This often means that the regulatory organisation should make clear that there are real and important social problems about which it will do nothing – not because it regards them as unimportant, but because Parliament has not given them powers. Ofgem was – is still – from time to time criticised for not promoting British national champions in energy, although the various Gas, Electricity and Utilities Acts which give Ofgem its powers are silent on the subject, as distinct from duties to promote competition or advance the interests of consumers (none of which is easily compatible with national champion ambitions). Similarly, criticisms of the FSA for not having a programme to counter social exclusion in financial services overlooks the complete absence of any such duty amid its (many) statutory duties.
Second, Parliament in giving powers to a regulatory organisation should not delegate essentially political or value judgments, as distinct from those which are essentially subject to technical or economic analysis. At Ofgem, for example, there was every reason to believe the regulatory organisation could devise rules and practices which encouraged efficient location of generating plant, or incentives for investment which enhance security of supply: it is well qualified to do those and other tasks. But it is not qualified, and should not be asked, to opine on questions of whether nuclear power is an essential component of Britain's energy policy, and an appropriate and economic response to climate change, or a technology whose immediate environmental benefits are outweighed by questions of very long term toxicity. Nor is this problem specific to energy. I would find it difficult to discharge the duties of Ofcom in relation to "applying adequate protection for audiences against offensive or harmful material" or "against unfairness or the infringement of privacy", and am therefore sympathetic to members of the Ofcom Content Board who have these unenviable tasks. But I have doubts as to whether they are judgements properly given to an independent regulator. In this, as in much else, the House of Lords Committee on the UK economic regulators, chaired by Andrew McIntosh, was right in its recommendation that "Government should be careful not to offload political policy issues onto unelected regulators".
Third, those responsible for taking regulatory decisions should be particularly aware of the need for continuity, predictability and due process. This places a premium on the need for full and thoughtful consultation on regulatory initiatives: consultation which aims not simply to count the number of comments, nor simply to weight them by the economic importance of those commenting, but which considers the arguments against the organisation's statutory duties. But it has much wider implications. You may have noticed that I have almost invariably in this lecture used the expression "regulatory organisation" rather than "regulator". This is not verbosity. It reflects a real and important difference. A mistake in early regulatory legislation was to confer powers on an individual, rather than an organisation, so that, for example, when I was appointed to the energy job all decisions, duties and powers were legally for me to take or discharge. It was a mistake because it personalised, trivialised and made subject to greater uncertainty than was either desirable or necessary regulatory decisions. It is a significant improvement that decisions, duties and powers are now given to corporate bodies, not to individuals – something corrected in recent energy legislation and got right from the start in the Financial Services and Markets Act which established the FSA. Not least, this provides continuity: when I step down as FSA Chairman in September, it will mean that one member of a board of 14 will change. I do not expect the FSA's policies or principles to change. This is very different from a world of individual regulators, when individual characteristics were analysed (and exaggerated), with all the uncertainty this caused when one regulator was succeeded by another.
It is clearly a strength of the Bank of England's responsibilities for monetary policy that there is such clear corporate governance in relation to the Monetary Policy Committee, and the carefully defined (and limited) responsibilities of the Court of the Bank towards the MPC. I think it entirely appropriate that the recent proposals to give the Bank a statutory responsibility for financial stability should be accompanied by proposals to bring greater clarity to the corporate governance of this important aspect of the Bank's responsibilities. Here, as elsewhere, clear corporate governance will enhance legitimacy.
Last, there is a critical element of legitimacy, namely the general question of how to ensure compliance with the principles and rules promulgated by a regulatory organisation, and how, within a general framework, to enforce those principles and rules. The general question of acceptance by the regulated sector is I think largely to be answered in terms of the issues I have discussed: the quality of the regulatory discourse, the predictability of regulatory decisions, the clarity of corporate governance within the regulator, the reality of the consultation process – in short, all these factors which contribute to a sense of fairness. For honest and well meaning firms and individuals this is what matters most. But an element of fairness must be that those firms and individuals who play by the rules believe that those who do not will be penalised, and that they themselves by observing the rules will not be disadvantaged. Effective deterrence of misbehaviour is an essential component of a fair as well as an effective regulatory regime.
Effective deterrence is not simply a matter of large fines. I am unimpressed by the assertion that, as some have claimed, the lower cost of capital in the US is due to the fact that the SEC imposes larger fines than does the FSA; or is a more enforcement-led organisation. For this claim to be other that simplistic it would be necessary to show that the cause of lower capital costs in the US was the fact that US capital markets were in fact less open to abuse than those in the UK. But there are many factors other than enforcement-related factors which affect the cost of capital: the depth and liquidity of capital markets are two obvious such factors. Even were the causal link to be established between the "cleanliness" of a capital market and the cost of capital in that market (and no such link has been established), it would be necessary also to show that the level of US fines or of US enforcement intensity actually resulted in "cleaner" US markets. But those who suggest that the size of US fines is more effective in preventing abuse have to accommodate within their argument the awkward reality that the analysts' scandal of the dotcom era or most recently the market timing abuses were abuses within the US market without equivalent in the UK; insider dealing is a substantial problem in both markets – as in others too. The assumption that higher fines lead to cleaner markets has to be proved, not asserted. As in much regulatory activity, the output, not the regulatory inputs, is what matters.
This is not to downplay the importance of enforcement. I regard it as an important weapon in the armoury of the regulatory organisation. But successful enforcement relies on many factors other than the size of fines: the probability of a breach being detected, the reputational as well as the financial impact of enforcement action, the scope for action against individuals rather than corporations, the impact of criminal rather than civil penalties, and the interplay between imposing penalties as a matter of course for breach of principle or rule and the inhibiting effect this has on candid and correct discussion of problems between regulated firm and the regulatory organisation. These are complex issues, where policy still needs to be improved. It is a subject to which the FSA will continue to pay close and growing attention. It is certainly more complex than a simple correlation between large fines and the cost of capital would suggest.
In all that I have discussed this evening, I have tried to indicate the variety and the complexity of what, following my wife who is a Shakespeare scholar, I have learnt to think of as the discourse of regulation. I hope that the issues discussed will indicate why those of us who work and are responsible for regulatory organisations find the work and responsibilities both difficult and – despite the criticism which is unavoidable – rewarding.