Q&A: Christian Clausen on bank capital, systemic risk buffers and bail-in debt

Systemically important banks will be required to hold more capital than their smaller, less-connected rivals – but no-one knows whether the systemic buffer has to be treated as an additional minimum, says Nordea chief executive Christian Clausen. That has big implications for planned recovery and resolution rules, he tells Duncan Wood

Three years after post-crisis reform efforts started in earnest, big questions remain about how the new regime will work, says Christian Clausen, chief executive of Nordea – not least how the new bank capital levels will interact with the embryonic recovery and resolution framework that is supposed to tackle the problem of too-big-to-fail institutions.

In theory, the capital rules reduce the chances of a big bank getting into trouble, while the recovery and resolution regime ensures taxpayers are not on the hook in any case. But there are a lot of details to work out, Clausen warns. The fourth Capital Requirements Directive (CRD IV), which implements the Basel III rules in the European Union (EU), means big banks will hold a minimum of 4.5% in common equity, with extra buffers driving the overall ratio higher – a capital conservation buffer of 2.5%, a counter-cyclical buffer of up to 2.5%, and a requirement for systemically important banks to hold anywhere from 1% to 2.5% on top of that.

What Clausen wants to know is where the systemic buffer sits in the capital stack, and whether banks will be allowed to eat into it, as is the case with the other buffers. That might sound like a minor, technical point but it has big implications. The recovery and resolution regime depends on two capital thresholds – when a bank dips below the first, the recovery rules kick in, potentially giving regulators the power to replace a struggling institution’s management team and board; when the second threshold is breached, the bank is deemed non-viable. At that point, remaining equity would be used to absorb further losses and the bank would be recapitalised by converting bail-in debt to fresh equity in sufficient amounts to catapult the bank back to health.

So, would a systemically important bank be deemed non-viable when it hits the 4.5% minimum capital level that applies to other institutions, or would the systemic buffer be an additional minimum, meaning resolution happens earlier? The same question applies to the recovery threshold. Depending on the answer, systemically important banks might be incentivised to hold an additional voluntary buffer on top of their existing capital. Holding excess capital could also reduce the cost of issuing bail-in debt.

Clausen raised these questions with the EU’s Economic and Financial Affairs Council (Ecofin) when it met in Copenhagen at the end of March. He says politicians have no answers at this stage. As a result, the optimal equity capital level for systemically important banks could end up being anything from 8% to 12%, Clausen says. “We really don’t know where the level will settle.”

I don’t think there has been a real discussion about where each buffer sits in relation to the others and what happens when you start to eat into them

The Nordea chief executive has also been president of the European Banking Federation (EBF) since December 2010, and he recently had a chance to air the industry’s concerns about another policy initiative – the expert group set up by the European Commission and chaired by Finnish central bank governor Erkki Liikanen, which is looking at whether the EU should follow the example set by the UK’s Independent Commission on Banking (ICB) and lay down rules about how banks should be structured. The ICB issued its recommendations in September 2011 – widely known as the Vickers report, after ICB chairman, John Vickers – and the UK government has backed its calls for retail banks to be ring-fenced from trading businesses.

Clausen says it’s impossible to know whether the Liikanen Group will reach the same conclusion – but it’s his impression that the initiative looks less like the start of a major new EU directive, and more like an attempt to form an opinion on the Vickers policies. And he says European authorities may end up concluding that the UK is breaching existing EU rules.

An additional concern is the liquidity regime that is part of Basel III and CRD IV – the liquidity coverage ratio (LCR) is one of two new ratios the rules introduce. It requires banks to hold enough liquid assets to survive a 30-day period of extreme stress, and is controversial in part because of the short list of assets deemed acceptable, which leans heavily on government bonds. CRD IV could broaden that list – and the Basel Committee on Banking Supervision plans to refine the standard by the end of this year anyway – but Clausen wants to see the list torn up and replaced with a criteria-based approach.

Risk: What has it been like to run Nordea during the crisis?

Christian Clausen: It’s been an amazing period. It’s been unprecedented by the standards of anything I’ve ever seen. I started in banking in 1979, so I have seen other major crises – not least the big one in the 1990s in the Nordic area – but this one has been more global, more systemic in nature, and the degree of uncertainty has been much higher. For the first time, we’ve seen financial markets being totally frozen, liquidity fully closed down. These are things we haven’t seen previously to the same extent, so it has been an incredible experience.

Of course, I’ve had the pleasure of running Nordea, which started from a very strong place and went through the crisis with quarterly returns on equity that never dipped below 8%, and no year that was less than 11%. But despite that, there have been periods where we’ve had to question things, including the business model. I think that is true of all banks – the whole industry is revisiting even very basic questions.

Risk: Sweden wants its banks to hold more capital than the Basel III minimums. What does that mean for Swedish banks competitively?

CC: In the short term, it’s not an issue. Right now, it’s a case of the more capital the better – and I think that’s true for all banks. In the longer term, we don’t know what impact it will have because we still don’t really know what the global capital minimum will be.

Recovery and resolution rules are one of the reasons for that. The regime will include capital thresholds that determine when a bank is in recovery mode – meaning actions can be imposed on it by regulators, including replacing management and the board – and when it has reached the point of non-viability, resulting in shareholders being wiped out and bail-in debt converting to fresh equity, recapitalising the bank. Those thresholds have not been set – and it’s not clear how they interact with CRD IV, because we don’t know whether the additional capital buffer for systemically important banks is one that can be eaten into, like the capital conservation and counter-cyclical buffers.

That’s important. If it’s another minimum, then the big banks will need to hold 4.5% as a start, and then a possible 2.5% on top as a systemic buffer – which means a minimum of 7%, and if that also becomes the level at which a systemic bank enters recovery, the industry will hold an additional management buffer above that level to ensure it doesn’t happen.

So, this regime may set a market practice standard that is higher than the minimum standards – and because we don’t know this yet we also don’t know how much bail-in debt we have to hold. Bail-in is going to be pretty expensive – and for that reason also the optimal capital level may be higher than the minimum ratio. We really don’t know where the level will settle. It could be 8%, 10% or 12% for a bank that holds a lot of bail-in bonds and needs to be able to recover.

Risk: Are policy-makers getting to grips with this?

CC: It’s a technical question that I don’t think politicians have decided yet – their discussions haven’t been that precise. I presented our view on this to the Ecofin meeting in Copenhagen and it was clear they had a very strong understanding of capital levels, but I don’t think there has been a real discussion about where each buffer sits in relation to the others and what happens when you start to eat into them.

I think it will be at least six months before we know this, and then we have to wait and see what happens on bail-in bonds. There will be a trade-off between the amount of capital you have and the price you pay to issue – the more capital you have, the lower the price of the bonds – so you can’t rule out the possibility that the optimal capital level is anything up to 12%. When I talk to other banks these days, the lowest ratio anyone mentions as a running level is 10%, and some people even mention higher levels.

Risk: How much more expensive will bail-in debt be than more traditional bonds? Some analysts are estimating an additional 100 basis points.

CC: There are two schools of thought. One says it will not be more expensive because bail-in bonds are transparent instruments for investors to buy – they know they will be bailed in when capital reaches 4.5%, they know existing stock will be written off, and they will be a shareholder of a well-capitalised bank, because a lot of bonds will be bailed in. So, the bank will have 10% in capital and the new equity investors will be able to unload their shares if they want to. The loss given default (LGD) on an instrument like that is pretty low when you compare it with today’s senior unsecured bonds – where, if something goes wrong, the bank could ultimately go bankrupt, leaving you with a claim on the estate. Assets will be sold, but you don’t know what value they will realise or whether there will be anything left – so the LGD on unsecured debt is very uncertain. You have to estimate it, but it’s very difficult to do.

The other school says obviously it needs to be more expensive because a 4.5% trigger is not zero – there might be a lower LGD, but the probability of default will be higher. Strictly speaking, it’s not default – it’s conversion into equity – but the point remains. Now, all of this depends on the sequencing of the buffers in the capital stack, and that hasn’t been decided. But, to me, it’s not likely that it will be very significantly more expensive than existing debt. Anyone predicting something in excess of a 100bp premium to traditional bank debt is putting it too high. But, again, it’s difficult to be certain.

Risk: What kind of feedback did you get at the Ecofin meeting?

CC: We got a lot of reaction afterwards from the governors and ministers. The way I presented our view was that you have the minimum 4.5%, then you have the systemic buffer, then the capital conservation buffer, then the counter-cyclical buffer, and a management buffer on top. That sequence sounds logical, and we know pretty well what the decisions will be when you burn through the counter-cyclical and conservation buffers – there will be no dividend, no bonuses, etc.

Then, when you break the 7% level and are into the systemic buffer, I think the most likely outcome is that part of the systemic risk can still be absorbed within the recovery plan but when you are between 4.5% and 7% then, if that lower level is the point of non-viability, you will obviously have some interim measures where management will be replaced and the board will be replaced. In other words, you can’t eat all of the systemic buffer, but you can eat some of it. That kind of thinking is shared by many people I talk to in Washington, at the Financial Stability Board and here in Europe, but it is absolutely not clear.

And again it comes back to the price of the bail-in bonds, because if you have some mandatory actions before the 4.5% trigger, the bondholders are even better protected.

Risk: Will it be a problem for Nordea if Sweden insists on higher capital levels than Denmark?

CC: No, it can be managed. But I still think people will realise that having different capital ratios in Europe is a systemic issue. Over time, banking will concentrate in areas with lower capital ratios, which presents a risk. Running a bank that is subject to different capital levels is certainly plausible – but having different levels throughout Europe, I find that less plausible.

Risk: Even if the only countries that go beyond the minimums are Sweden, Switzerland and the UK?

CC: Maybe it will take some time, but if there is a significant difference then banking will concentrate over time in countries such as Germany and France, where capital levels are lower – you could see banks start booking more loans there, doing more business there, moving their headquarters, maybe re-domiciling and so on. That will gradually happen. Then there’s the US, which will probably end up with a capital minimum of 7–8%. It will cause concentrations of banking.

But this is some years away, because as long as the markets remain uncertain, high capital levels are not a disadvantage – I think every bank will try to go higher than they need, to get cheaper funding and show the market they are strong.

Risk: Is Nordea compliant with Basel III’s LCR?

CC: Yes, we are – we have the value at more than 100%, and the composition almost matches the Basel III definition. It will be a very small extra cost to adjust further. Of course, the Basel Committee on Banking Supervision is currently reviewing the details of the ratio. We don’t know when and how, but we know the committee is going to come out with adjusted parameters.

Risk: Do you think that will result in eligible assets being decided via a criteria-based approach?

CC: Unfortunately, I think the Basel Committee will be rigid on this. But obviously they don’t make the laws, and when legislators step in, it could change. The US has said very, very clearly that it will have its own definition of liquid assets, and this means there will need to be some adjustment between the EU and the US.

I think bankers and regulators alike recognise that some calibration and redefinition is needed to avoid weird side-effects, so if the Basel Committee doesn’t do it, lawmakers will. But I don’t expect it to make a world of difference – it will still be a harsh regulation that will be costly to the banks.

Risk: What kind of side-effects?

CC: Concentration risk, specifically. If you want banks to hold liquid assets, but you want to avoid concentration risk, then you can include some alternatives to government bonds in level 1 – assets that are also very liquid and would reduce concentration risk. That’s another approach and is very logical. There is a reality out there, in terms of the types and amount of available assets, and the relative levels of liquidity – and the LCR needs to be adjusted to reflect that reality. Also, there has to be some correspondence back to the assets that are defined by central banks as being eligible. If a central bank will accept them, then in some sense they are liquid.

Risk: CRD IV asks the European Banking Authority to look at adding some other liquid assets to the list – and also to reassess the existing treatment of government bonds. What do you expect to happen there?

CC: My fundamental view is that we need to change it completely to a criteria-based approach – it needs to be diversified, it needs to be liquid, and we don’t want systemic risk in the liquidity buffer – so, to me, we need to think it through in a different way. If that’s not possible, and we have to work with the existing system, then it has to be recalibrated to ensure government bonds have less importance and other bonds have more – and this notion of tying it to central bank-eligible assets is important.

Risk: Apparently, you had the chance to talk to the Liikanen Group. Is that right?

CC: Yes, I met the Liikanen Group – we sent them some material in advance. I then presented the Nordea business model to them and also represented the views of the industry more broadly on behalf of the EBF.

Our view is they should not impose new structural changes similar to Vickers or Volcker – they should use the existing regulation that is now being built, because that will fully take into account the risks you really want to mitigate. What Vickers wants to do, in terms of ring-fencing – and what Volcker does, in banning proprietary trading – the same things can be achieved within the existing framework for capital and risk, so there’s no need for a further round of change.

But I have the feeling that the Liikanen Group itself has a similar starting-point, and that was supported by a statement made recently by Michel Barnier [the European Commission’s commissioner for internal markets and services] in which he said he does not support a Vickers approach in Europe. That’s my feeling right now, but where they will end, nobody knows, and of course the growing trading losses announced by JP Morgan are not at all helpful here.

Risk: Prior to the Liikanen Group being set up at the start of this year, there didn’t seem much appetite in Europe for structural reform of banks. Did it come as a surprise to you?

CC: Yes, I was somewhat surprised. I shared the view that it has not been seen as a big issue. It probably came from the fact that the UK is working on Vickers and ring-fencing, and it must be difficult for the EU to stand by and watch a member state working on something that is a radical and interesting solution without at least investigating the options – and also investigating whether it can be done within EU rules, because that is the other thing they will come out with, of course. The UK can’t do something that is against EU regulation – that’s part of it.

Risk: Where is the potential conflict?

CC: It could easily come out of several directives. The capital requirements in Vickers are very different from CRD IV both in terms of minimums and also as regards other loss-absorbing levels – then you have all these restrictions on the relationship between retail and investment banking – so, I don’t know that it is in conflict, but it’s also not obvious that it’s not in conflict.

Then, the second issue for the EU is, what if other countries wanted to do something similar – could you allow a free-for-all? I don’t know this, but my guess is they decided they had to have some kind of stance, so they’d better investigate. And it’s noteworthy that it’s being done by this special high-level committee, rather than the usual format they would use to prepare a directive.