Speech given by Martin Taylor, External Member of the Financial Policy Committee, Bank of England. Based on remarks given at the International Association of Credit Portfolio Managers conference, London
It is both a pleasure and a privilege to talk to you this morning. These are strange times in the world of credit, and you must all have strong nerves to be in the business of managing it. When I was first involved in finance, an unquestioned credit was defined as a borrower who would undoubtedly pay you back all that you had lent to them. Recently it has come to mean a borrower who will undoubtedly pay you back a little less than you lent. As a sub-investment grade speaker I shall try to give statistically-adjusted value today.
I've been asked to talk about regulation, which is of course a vast and complex subject as well as a controversial one. I'm going to give you a rather personal slant on regulatory developments, which I've observed from different standpoints at different stages of my professional life. In the 1990s I was a bank chief executive on the receiving end of the Bank of England's humane but penetrating supervision. Later, having skilfully arranged to be somehow absent during the financial crisis, I returned to the fray as a member of the UK's Independent Commission on Banking, chaired by Sir John Vickers and known to everyone except him as the Vickers Commission. It was our job to propose the regulatory response to the crisis that might be appropriate for the UK – this in 2010-11, well before the smoke of war had begun to clear from the international regulatory battlefield. And now, as an external member of the Bank of England's Financial Policy Committee, I struggle with the thorny problems of macroprudential policymaking, which very much include the bank capital and too-big-to-fail agenda. For the avoidance of doubt, everything I say today represents an entirely personal view.
When I was a bank CEO I believed we had plenty of capital. Bank CEOs tend to believe this until it is clear to everyone that it isn't true. In particular I was conscious that the regulatory capital prescribed by the supervisors, based in those days on Basel I, exceeded by some way the economic capital we calculated for ourselves as being the desirable quantum. We indulged in various portfolio adjustments to try to bring these two important numbers – both highly conceptual, not to say imaginary – into line with one another. I now understand that my serene confidence in the state of our capitalisation may have been unwarranted, though in the mid-90s all the big banks were far better capitalised than they came to be ten years later. What I have also learned over the last 20 years is that if the risk management skills of the banks (which I understand this conference, most laudably, is aiming to improve: thank you so much) had sustained the same high level as their prize-winning abilities in regulatory arbitrage, the world would be a safer place, and we might all have been spared a lot of trouble.
When the Vickers Commission met for the first time in June 2010 we were contemplating the ruins of a system. The system was in intellectual as well as financial disarray. Different countries were responding with different degrees of urgency, mostly depending on how badly their economies had fared during the crisis, which in turn was highly correlated with the balance sheet size of their banks relative to their GDP. Britain was somewhere near the top of this melancholy table, below Iceland and Ireland, but right up there with Switzerland. Academics were having a field day, wheeling out recommendations – enticing but not always entirely practical – for the complete abolition of the banking system as it had existed for many decades. Meanwhile the banks were saying, in operatic chorus enlivened by the odd coloratura solo, that the crisis had been a one-off, freakish event, which essentially required no corrective measures; that any increase in regulation would destroy the financial sector in the UK; that it would prevent them funding what was then a non-existent economic recovery; and that if we came up with something they didn't like, they would move out of the country. A cover version of this last track may be on your iPod again right now. In 2011 I came to worry less about this issue when I learned from friends at the Swiss National Bank that many Swiss banks were simultaneously threatening to move to London. For what it's worth, and speaking as a former banker, I don't believe regulation alone is likely to prompt bank re-domiciling – apart from anything else, the rules are not that easy to escape.
Vickers recommended both that the banks should hold far more capital – and capital of better quality – and that measures should be taken to ensure that in future bank collapses bondholders, and maybe other providers of finance, should be on the line before taxpayers. These two crucial ideas form the foundation of the international architecture being developed at Basel under the auspices of the Financial Stability Board. A fair amount of important detail remains to be settled, but the shape of the cathedral under construction is now pretty clear. It is good news for citizens of the advanced economies that the FSB has had the determination and courage to go a little further than, back in 2011, some of us feared it might. Since then, too, stress testing has been developing, most strikingly in the US but subsequently elsewhere, and a gross leverage ratio with teeth – to complement the risk-weighted measure – is finding its way into the rule-books (this was another Vickers recommendation).
Vickers also proposed – and this has since been voted into law by parliament here – that the retail banking activities of the large banking groups should be ring-fenced from the rest of their business, principally, that is, from investment banking. (The EU committee under Erkki Liikanen suggested a policy that resembled a photographic negative of Vickers, involving the ring-fencing of investment banks.) At the risk of boring those of you who have no interest in the minutiae of British bank regulation I should like to dwell for a moment on this question. The reason I want to do so is that a number of voices have recently been raised arguing that ring-fencing is redundant, given other supervisory developments, and calling for the law to be repealed.
Those who argue in this way tend to have some professional connection with the banking industry, but let us be charitable and assume that they are public-spirited and driven by purely philosophical motives. Essentially they say that the banks are now, or shortly will be, so well-capitalised that the additional protection afforded by ring-fencing is unnecessary. They claim that the undoubted private costs of ring-fencing (to the banks) outweigh its public benefits.
This argument starts from the premise that the ring-fence was invented to make banks safer. Bankers have been saying since it was first proposed that the ring-fence would not effectively increase resilience, arguing – as people are arguing again – that retail banks could be just as dangerous as investment banks. I broadly agree with this last point. But to argue that making banks safer is the principal or even only aim of the ring fence is simply wrong. What I am about to say has been said very many times, especially by former members of the Vickers Commission, but so many people in the industry have either failed to understand it or have chosen not to that I shall permit myself to say it again.
The establishment of the ring-fence had three objectives. First, it was intended to make resolution easier, to allow the authorities in a crisis to pluck out the critical elements of a failing bank – exactly what could not be done in 2008 when RBS failed. Resolution, which bankers prefer not to think about most of the time, is an absolutely crucial element of the post-crisis architecture, since (as Lehman Brothers convincingly demonstrated) conventional bankruptcy proceedings don't work for banks. I'm happy to say that this is an area in which major strides are being made.
The second aim was to break the mechanism by which the banking industry enjoyed an implicit subsidy from taxpayers. As long as governments stood behind a bank's domestic business with its insured deposits and its access to the payments system, and as long as that domestic business was mixed in with the investment bank in the sense of sharing the same balance sheet, then effectively the taxpayer had no choice but to stand behind the investment bank too. (The credit markets, ladies and gentlemen, understood this implicit subsidy perfectly.) So the ring-fence was designed to isolate the retail balance sheet and thus to sever this link and subvert an irresponsible business model that had proved very profitable for the bankers and very costly for the public.
Indirectly, the disabling of this dangerous practice should indeed contribute to making banks safer. That is the third objective. The ring-fence structure, incidentally, goes far beyond the single issue of outlawing the use of insured deposits to fund trading activities which the Volcker rule in the US rightly aims to achieve. I believe that the fuss about the ring-fence at the moment, far from demonstrating its redundancy, shows that its forthcoming introduction is proving effective. It is beginning to do its job, and of course not everybody likes that.
This illustrates a particular instance of a more general phenomenon. Wide-ranging regulatory changes are having structural effects. This should surprise no one, and in general I feel it is quite wrong to fret, as so many seem to be doing, about the so-called unintended consequences of the new rules. If we feel that some, such as the reduction in trading liquidity, may bring problems in their wake, let us address them individually. Let's not forget – before the crisis, the banks were quite simply in the wrong place; they are now feeling their way towards what we hope will be a safe, decent and sustainable model for finance. In some cases this will inevitably require violence to be done to established ways of proceeding.
Vickers stopped short of recommending that the banks should be broken up; we recognised that there might be benefits in portfolio diversification between different kinds of banking businesses. But we certainly understood that the new rules would force banks to examine their business models and their portfolios. All developments in banking between the 1970s and the crisis had tended to increase the homogenisation of bank balance sheets. Banks had become soups – opaque, oceanic soups of assets, assets whose correlations with each other were imperfectly understood. You can't carve up a soup or an ocean. But moving to a more crystalline industry structure may help us handle too big to manage as well as too big to fail. Even if some of the little frigates that eventually emerge from the industry re-shaping turn out to be individually riskier – because less diversified – than some of the aircraft-carriers, the consequences of their sinking are unlikely to be as serious.
Alongside attacks on the ring-fence come more general calls – particularly strident in the last couple of weeks – for the regulatory pendulum to swing back. Clichés are a poor guide to policy, but the mention of big swinging pendulums inevitably introduces the question of cyclicality. And for those of us on the FPC, counter-cyclical policy intervention is a critical objective.
It's not that easy, however. Charles Goodhart and others have pointed out that macroprudential policy, while intended by its very nature to be counter-cyclical, generally turns out to be the complete opposite. Put simply, we tend to tighten regulatory policies straight after a financial crash, when the economy is so weak that – all else being equal – we should prefer to loosen them. But in 2008/9 policymakers had little choice: regulation was at a maximally permissive setting – it had only one way to go. Of course everyone could see that it would not be straightforward for the banks to support the economy while re-building capital. That's why they were given plenty of time to move to the new capital standards, while monetary policy has been and remains extraordinarily accommodating. At last we may be working our way through this difficult phase, during which the direction of macroprudential policy has been broadly pro-cyclical. Helpfully, an important force has been working in the other direction: as banks have become better capitalised – a process that would have happened a lot faster, of course, had past misdeeds not come home to roost in the form of tens of billions of misconduct penalties and customer redress – their ability to extend credit has visibly strengthened. Weak banks simply don't lend.
While the banks build capital levels to the targets required by 2019 – and many of them, pushed by shareholders not to drag their feet, are there already – the economic recovery is maturing. At some stage, then, that capital build stops being pro-cyclical, turns neutral and then counter-cyclical. Bodies like the FPC are now supposed to increase bank capital by engaging the Countercyclical Capital Buffer as the cycle moves towards its peak, in order to be able to release it and thus support economic activity on the way down. In this cycle so far it has been unnecessary to raise the CCB, partly because banks have still been facing the requirement to add capital towards their 2019 targets.
We don't precisely know where we are in the cycle, of course, which is why counter-cyclical policy is so tricky. And the post-crisis narrative in Britain as elsewhere is that the recovery is in its very early stages. But it's been a long time since 2009, and this economic cycle, restrained though the recovery was at first, may well be closer, measured in time, to its peak than its trough. Potentially, we're reaching the point at which incremental bank capital-building will begin to have a desirable counter-cyclical effect. So calls from banks to swing back that pendulum and reduce regulatory requirements are doubly to be resisted: resisted once because we are still on our way to the new paradigm, resisted twice because – if anything – we're reaching the point of the cycle where macroprudential policy should be tightened in any case. The policy question here for me – and it's an open question – is at what stage we may need to tighten it over and above what's been happening automatically as the banks have transitioned to the new regulatory rules. But loosening now would certainly make no sense.
Regulators are often accused of fighting the last war – another cliché to put alongside the pendulum. To which my reply would be: the war may be over but the details of the peace treaty have not yet been fully worked out, and the territorial changes consequent upon that peace treaty – by which I mean the structural changes, spin-offs and re-shapings in individual organisations and the system as a whole – are by no means complete.
Alongside attacks on the ring fence and calls for the regulatory pendulum to swing back, it has become fashionable to worry about risks displaced and dispersed outside the banks by the greatly heightened regulatory scrutiny to which the banking system has been subjected. Before the banking crisis we all congratulated ourselves that risks had been widely spread throughout the financial system. We then came to understand that many economic agents held risks that they understood very poorly; that these risks were often wrongly priced; and that in many cases it made no sense for those who held them to be exposed to them. To compound the calamity, some of the apparently "dispersed" risks were actually backstopped by banks, or had been mis-sold, so that in reality they had never left the banking sector, whatever the banks' balance sheets may have suggested.
It's important to distinguish here between two types of risk. There are what I might call natural risks – the inherent risks arising from fluctuations in economic activity and the fundamental uncertainty of outcomes that underlie every financial asset, whether it's equity, debt or hybrid, and whether it's held on a bank's balance sheet or in a portfolio. The supervisor may be concerned about the distribution of these risks, but can hardly worry about their existence. They accompany our lives and are the source of our prosperity – oh yes, we on the FPC are very concerned about prosperity.
Then there are synthetic risks, created for instance by leverage or by maturity mismatch, which serve to magnify the natural risks. These types of transformation can help the economy to function more effectively. Financial institutions take on synthetic risk in order to increase their returns, sometimes by holding risky assets in risky liability structures. Banks still do this more than any other type of institution, and it is not foolish for supervisors to concentrate on them. Indeed, the biggest dangers posed by so-called shadow banks may be dangers that their failure might pose to the banking system.
One of the most striking new structural features of the post-crisis banking architecture is that more bank creditors will be at risk in the event of failure. When Vickers looked at this issue five years ago bail-in bonds and convertible capital notes (CoCos) appeared to be alternative solutions to the same problem. They now appear to be complementary, with high-trigger CoCos, which have become the norm in the London market, supplying going-concern capital for banks – that is to say, a bank may be expected to survive the triggering of a CoCo, even if its management is unlikely to – while bail-in bonds, probably issued by a holding company so as not to compete for precedence with deposits, are there to save the taxpayer from financial exposure when a bank fails. It follows that the holders of these instruments – credit portfolio managers, on the whole – have an important role to play in the new structures. At last the supervisor has an ally in policing the excesses of bank managements. (Previous potential allies, equity holders and non-executive directors, proved ineffective in the build-up to the crisis: indeed, they mostly seem to have been fighting on the other side.)
I talk to a group like this about credit matters with the greatest timidity. I am sure you are good citizens and desire to exercise exemplary scrutiny. But I wonder whether you will flip – like the holders of European sovereign bonds before 2010 – from believing all issuers equally safe to thinking many equally precarious when the sky next darkens. I worry that CoCos may be subject to potentially destabilising manipulation by convertible arbitrageurs, which is why I think we should be careful how large a role we allow them to play. I note that different kinds of investors are inclined to believe contradictory things. So at the same time that some equity investors claim that banks, in the interests of safety, are now forced to hold so much capital that they may struggle to earn proper returns, some debt investors can be heard to say that CoCos and bail-in debt are dangerous instruments to hold since banks are so precarious. I don't believe either of these propositions, as it happens, but I find it very difficult to see how anyone can believe both at once.
To sum up: this is a crucial time for the new international order in bank regulation. We are close to agreement on new standards that the industry, in the UK at least, is not too far off meeting. Four years ago that would have seemed a highly desirable outcome but quite an unlikely one. It's good for our economies, and it will turn out to be good for the financial industry over the next quarter-century. At the same time the emergence – well, they never went away – the increasingly shrill emergence of voices calling for a regulatory softening is both structurally wrong and conjuncturally wrong. It remains the ungrateful job of the supervisors to save the banks from themselves. The shortness of human memory span and the speed with which we forget the ghastly misjudgements of the recent past: these are the enemies, the unresting enemies, alas, of financial stability.