Ed Sibley is the Deputy Governor of the Central Bank of Ireland with responsibility for prudential regulation. It’s his job to make sure the banks don’t go bust again. 

The Central Bank took the job of prudential regulation back in-house after Patrick Neary’s catastrophic term at the Irish Financial Services Regulatory Authority from 2003-2009.

The main tool the Central Bank of Ireland has at its disposal is its implementation of eurozone-wide capital requirement rules. Capital requirement rules dictate how much money a bank has to set aside to cover potential losses. The more risky a bank’s loans are judged to be, the more capital it has to set aside. The more capital it has to set aside, the less profitable it will be. 

The capital requirement rules come down from the ECB headquarters in Frankfurt, and in theory they are the same across the eurozone. But in practice, the rules are implemented separately by the eurozone’s 19 national central banks. And interpretations of the rules differ widely. The Central Bank of Ireland says loans to Irish people are among the riskiest in the eurozone, and makes Irish banks hold more capital against those loans than any other eurozone country. 

Why should you care about bank capital requirements? They matter for three reasons.

First, they’re a reason KBC and Ulster Bank are leaving Ireland. Both banks said they’re getting out of Ireland in part to free capital “trapped” by the Central Bank’s rules. KBC and Ulster Bank were followed by the giant Italian bank UniCredit in May, whose Irish unit is to shut down after 26 years in operation.

Second, they’re a tax on borrowing. The extra capital Irish banks are asked to hold adds about half a percentage point to the cost of an Irish mortgage. Half a percentage point, over the life of the typical mortgage, comes to about €29,000 in extra interest payments.

Third, they’re a drag on bank profits. When the Central Bank of Ireland’s says Irish mortgages are about twice as risky as those in Europe, the result is that the banks’ profits on each loan are cut in half.

The man on the street might not care much about returns for bank shareholders. But banks are now so unprofitable that they can’t get private investors to fund them. The only remaining source of funding for our unprofitable banks is the state. 

To the extent that they’re preventing banks from making a profit, capital rules are having the effect of pushing bank borrowing onto the government’s balance sheet. This is precisely the opposite of their intention.

In this conversation, we discuss:

This conversation has been lightly edited for clarity.

*****

The drawbacks of capital requirements

Sean Keyes: The benefits of higher capital requirements are straightforward: they make a financial crisis less likely. What about the other side of the equation? What are the costs of capital requirements?

Ed Sibley: We certainly try to take a pragmatic view of this. I mean, fundamentally, why is capital there? It is safeguarding financial resilience and safeguarding stability, but it is really about protecting consumers and protecting the depositors as well as being a key driver of risk management within a bank. So there are lots of aspects to think about in terms of the appropriate level of capital, including what are we capitalising for.

SK: Another way of putting the question might be, why not raise capital requirements further?

ES: Ultimately we operate in a system where it’s a market-based economy. The banks are there to take risks and generate profits. So in that model, as a sustainable business model, a key aspect of that is the need to generate sufficient return on its equity. By its very nature, the higher the level of capital required to be held, the more challenging it is in terms of return for that business. 

But I don’t look at it through a single lens. From a system-wide, and also from an individual bank’s perspective, there are four things that we need to be considering. 

Firstly, the adequacy of financial resources. So that’s both capital and liquidity, including into the medical, plausible stress scenario. 

Secondly, the sustainability of the business model. So is this business likely to generate returns on capital accreted over time, including good times and bad times?

Thirdly, is it well-run? So: governance, risk management culture.

And then fourthly, if the firm gets into difficulty, can it recover based on its own steam? And if it can’t recover, can it be resolved without significant externalities or cost to the taxpayer? 

So those are the four aspects that we apply to thinking about supervision. We don’t look at one in isolation over the other. So yes, you could have banks capitalised at 50 per cent CET1. And they would consequently be less likely to fail. But they would also be much less likely to have a sustainable business model. And also, probably that would have some effect on their risk management techniques as well. So it’s not that there’s one measure. But it’s looking at it in a pragmatic way. And ultimately, what we care about is the functioning of the system as a whole. 

“Is it any surprise to me that certain banks might be complaining about capital levels or the capital regime? No, not at all.”

SK: We don’t want financial crises. One big reason we don’t want financial crises is that losses get pushed onto the sovereign balance sheet, and that’s bad. But if banks can’t generate an economic profit, then the inevitable long-run consequence of that will be all bank lending getting pushed onto the sovereign balance sheet. Which is a different route to the same destination. 

Across the eurozone, and in Ireland, banks haven’t generated an economic profit for 13 years. Is that not a hint that maybe the Central Bank of Ireland and the ECB have got the balance wrong?

ES: It’s a really good question. And it is worth thinking about in that context. But I think we have an ability to look beyond what’s happening in Ireland, because of the way the European system works, but we also have the benefit of being able to look beyond Europe. So we can look into the UK, we can look to the US. 

Now there are lots of issues affecting European banks and their relatively low-level returns on equities, on average. Some of those are consistent both over time, but also across jurisdictions. So an obvious one would be low-interest levels. Another one would be that a lot of banks depend on interest income as opposed to a more diverse model. So those things are similar across Europe. 

In some countries, typically the bigger ones, they would say they are over-banked. So they might say there’s too much competition, which is going to drive down margins. And in other countries, there’ll be other reasons for the lack of profitability. In the case of Ireland, there’s the continued drag from legacy issues, historical underinvestment in IT, a high cost basis and so on.

Banks leaving Ireland

“There are real challenges for the European banks, generally, but Irish banks specifically.”

SK: There are surely are some things about the current setup that give you pause, though. The banks aren’t making an economic profit. They’re leaving the country in increasing numbers. We’ve had three in the last two months. 

ES: I think we’re not in the right place, in terms of issues in the banking system. And they clearly need to continue to be addressed. There are real challenges for the European banks, generally, but Irish banks, specifically, need to be focused on. 

I don’t think the answer is looking for a special case for Ireland in terms of the application of the European rules. And really, I think that there are lots of factors that are driving a particular decision, like retrenchment.

National banking that came out of the financial crisis led to a very significant reduction in competition in Ireland. So there was an immediate three, four, five-year impact. We now have in 2021 two particular decisions – one that has been made, one that may well be made. I think they’re negative from a competition point of view. That is of concern. I think the role of the Central Bank is to be trying to strive to address, as much as we can, the underlying issues, not looking to have under-capitalised banks operating in Ireland. 

The effect on mortgage interest rates

SK: What do Ireland’s RWAs add to a mortgage interest rate, relative to a typical European one?

ES: It will vary a little depending on the type. I personally think it’s about 50 bps. (0.5 per cent).

SK: 50 bps is significant, though. Throughout a typical mortgage, it might be €30,000. 

ES: Yeah, but… I agree with that. So I think that there’s, and again, it’s something that we’ve been pushing on, and we’ll probably continue to do so. That’s very significant. And one of the reasons that I would continue to push people to look at what they’re paying for in terms of it. 

SK: But competition isn’t going to fix it, is it? You’ve got this if you’ve got higher capital ratios across the board for every lender. That’s just something you have to live with. Right? 

ES: I disagree with you on that. If we had this conversation five years ago, I’d agree with you. I’d have said, ‘wherever you are, whatever products you’re on, they’re all pretty much the same. There’s very little difference between them.’ There is a very significant difference now. On mortgage rates, there’s a significant enough range, significantly more than 50 bps, between different products that are available. 

Anyway, fundamentally, I think there are definitely issues there. There are definitely issues at a European banking system level. Broadly speaking, in terms of what I can see, I think we’re in about the right place. And there are actions that we have taken, and that will continue to take that will look to continue to drive down the underlying risk. Which will have an impact on risk weights. And there is also action being taken on the other side, in terms of reducing the impact of internal models, which I think won’t have a negative impact on the retail banks here in Ireland, that might well increase risk weights elsewhere.

Why the rules haven’t changed

SK: A lot has changed in the last 10 years. So why hasn’t there been much change in the risk-weighted asset regime? The loan book is one thing. It has been gradually re-risked, bit by bit. But there’s more qualitative stuff as well: there’s a better regulator, better governance, better credit underwriting standards, more broadly based lending across the economy. The new macro-prudential rules are another big one. All of these things have made the system safer. Yet as far as the capital requirements regime is concerned, the banks are as risky as they were ten years ago. 

ES: Firstly, risk rates are trending slowly downwards. Because of the composition of the book. That is having an effect. 

There are other factors at play in terms of thinking about the riskiness both at an individual firm level, but also at a macro level. There are a couple of different lenses to look at this, in terms of are we off the charts? What are the broader macro risks? And there are macro risks that are probably more acute in Ireland than in other jurisdictions. The volatility of the Irish economy, I think that the challenges in the overall housing market are also relevant here.

“I don’t think the answer is looking for a special case for Ireland in terms of the application of the European rules.”

SK: But if you just look at Ireland in isolation – surely it has changed for the better in the last nine or ten years? Why hasn’t that change been reflected in the risk weightings?

ES: There are lots of things that I would hope have improved over that period. But one can also look at it from the other direction. So when one thinks about it from a stress-testing perspective, do we end up with massively overly capitalised banks in a plausible stress scenario? And the answer is no. So broadly speaking, from a high-level perspective. I’m certainly satisfied that we’re in the right ballpark of appropriateness in terms of capital levels. 

We totally recognise that there isn’t a free lunch there. There are associated issues with that. I do think that the direct causal impact effect of higher risk weights on the mortgage side is – that’s the only reason or the main driver for mortgage pricing to be higher in Ireland – I think that point is overstated. 

We’ve talked a little bit about it before, where if one thinks about the amount of capital that a bank is required to hold for any given mortgage, the necessary return on equity for that capital is being held in terms of a mortgage, actually comes to a fraction of a percentage point. And so there are other factors at play in terms of mortgage pricing. There are other issues in terms of costs, and one of the things we are striving to do is to try and look at some of the fundamental issues that are at play, including long-term mortgage arrears, and continue to strive to address them, which ultimately one would hope will make Ireland a more attractive place for banks and non-banks to lend.

SK: The Central Bank of Ireland is an Irish institution, but it’s a eurozone institution too. It sits between the two. So I’d like to understand to whom are you responsible? Is it the people of Ireland? Or are you responsible to the European Union?

ES: So I serve the people of Ireland. We’re absolutely focused on delivery for the people of Ireland. And we are held accountable in a number of ways, but formally, through our relationship with the Minister of Finance, and publicly through our engagement with the Finance Committee of the Oireachtas. And also other arrangements. But there is no confusion or doubt in our mind – we are here to serve the people of Ireland. 

Now, we do that in a number of different ways, including how we engage within the European Framework, both from a monetary policy perspective, but also in terms of financial regulation and financial supervision. I think that that is not a perfect system by any means. But overriding this, I think it is a positive that we are active and positively contributing to the European dynamic. There is a definite good there. 

And I would also say that’s much less talked about. But it is good for the Irish economy, in terms of the outsized nature of the Irish financial system, which is completely dependent upon continued access to the euro from Ireland, and the reputation of the Irish financial system. 

SK: Does the opacity of the capital rules make it difficult for a big bank, like BNP Paribas for example, to come in and acquire an Irish bank?

ES: There’s a rich history of the political economy of regulation and the commentary on how regulated entities are being treated at any particular moment in the economic cycle. So I think we’re at an interesting point in that historical trend. And is it any surprise to me that certain banks might be complaining about capital levels or the capital regime? No, not at all. BNP Paribas or any European bank will be operating to exactly the same required regulatory requirements, exactly the same approach to supervision as is currently the case for Bank of Ireland. 

SK: Should things continue as they have, with the same progress in the wider economy, lending standards, etc.? When do the bad years wash out of the back of the model? When can we look forward to Irish risk-weighted assets converging with the rest of Europe, subject to everything going fine?

ES:I think there’s a really important point that there will be movement both ways. In the scenario you described, it would be happening in both directions. So if you like, the graph is moving one way for the Irish banks, and the other way for the rest of Europe.

Why interpretations differ

SK: You say you don’t want Ireland to be treated as a special case. But when I talk to bank managers, they say LGD (loss given default) is the most important factor in their risk weightings. And then if you compare Ireland’s loss given default to Ireland’s risk weightings, Ireland is a massive outlier in the eurozone. And so that speaks to a special treatment in a sense – there’s not a uniformity there. There’s a big spread between Ireland and other eurozone jurisdictions. It seems like the rules have been implemented quite differently in a place like Spain compared to Ireland. 

Perceived riskiness of assets versus loss given default in eurozone countries; Ireland is on the top right. Source: BPFI

ES: We’ve talked about the importance of loss given default, but the probability of default (PD) is also important. When we’re talking about multiplication effects, then having higher LGD and higher PD – the mathematical relationship will explain some of the effects that you’re talking about there. Today, the probability of default for Irish mortgages is higher than the European average. So we’ve got higher PD, as well as higher LGD. 

But I’m very confident that we have gone at this in a consistent way across Europe. And one of the best things that that single supervisory mechanism has undertaken in its five or six years of existence has been the targeted review of internal models (TRIM), which has sought to address the inconsistency in how internal models are looked at across Europe. With a very high degree of governance review at the centre, to make sure that there aren’t jurisdictions that are either over or under egging it. 

SK: Just on that point – after the TRIM, the spread in RWAs between Ireland and other jurisdictions increased, didn’t it? So it hasn’t been the case that the TRIM has fixed inconsistencies in how the rules are applied. What’s happened is that the results have gotten less consistent, because Ireland’s RWAs have moved even further away from other European jurisdictions.

ES: In some jurisdictions, it resulted in reductions, in some cases it resulted in increases. I would go back again to – why is that the case? Is it an issue with the application of the rules and requirements in terms of banks’ models? Or is it an issue with the fundamentals? As I touched on earlier, we’re pushing hard on trying to address mortgage arrears. We’re dealing with people who are 10 years in arrears. A significant number of mortgage accounts are in deep, deep arrears. We haven’t, as a system, within the individual banks, outside of the banking system, being able to address that. So that was one particular important issue. And there are other issues at play. So I think where we should be focusing our efforts is addressing fundamental risks rather than looking for special treatment. 

Opaque rules

Ed Sibley: “There are judgments and assumptions in there.”

SK: Looking at this from the outside, there are three hints that capital requirements are too high in Ireland. The first is that, when plotted against loss given default, Ireland’s RWA’s are a big outlier compared to other European countries. The second is that the spread between Ireland’s RWAs and other European countries has gotten wider in the last five years, despite Ireland’s economy and lending standards improving a lot. The third is that three major banks have left Ireland in the last few months, and very few moved here after Brexit. You may say that your models justify this. But you won’t share those models with anyone. If Ireland really is riskier, why doesn’t the CBI explain its reasoning? 

ES: Fundamentally the reason the capital requirements, the risk rating, is higher in Ireland than European averages is connected to risks. Standardised rules are transparent to everyone. It’s the firm’s own modelling, not ours for the internal models. They follow a prescribed set of rules and they deliver output, in terms of risk weights, that is connected with the underlying risks. There are judgments and assumptions in there. 

SK: But there are important parts of the model that you don’t share with the banks and the country. The probability of default is one, recovery is another. You don’t share how they trade off with each other. And that opacity has consequences.

ES: We review the bank’s own internal audit. So it’s not that we got a secret formula that we’re applying and imposing on the banks. We have a set of requirements that are consistently applied regularly at regulatory and supervisory level across Europe. Banks can undertake their own modelling of risk, which provides the risk weighting. Now, they are subject to challenge. So they are challenged by us. And if we feel as though they’re missing something, or underweigh somewhere, we will change back and ultimately, we do have to improve the models. 

I’d like to make a couple of other high-level observations. Fundamentally, I think we’re broadly in the right area. And given the risks, the potential stresses, the trade-off that we talked about, and the need to be pragmatic, I think we’re in and around where we should be.

“A significant number of mortgage accounts are in deep, deep arrears. We haven’t, as a system, within the individual banks, outside of the banking system, being able to address that.”

SK: I think anyone would concede that Ireland has lower recovery rates on mortgages, and it has a knock-on impact on riskiness, capital ratios, ultimately interest rates, consumers, all that sort of stuff. But if we want to have that conversation in Ireland about whether we should have greater enforcement of property rights, and lower interest rates, at the expense of more people having their homes taken away, it’s difficult to have that conversation when the CBI doesn’t make explicit how that feeds into your risk models. We’re told that it matters, but we don’t know how much it matters. And if we don’t know how much it matters, we don’t know how much it’s costing us. If we don’t know how much it’s costing us, we can’t have that debate in the country. 

ES: I totally take your point about whether we could be more transparent. I’m certainly happy to reflect on that. Like I said, I really welcome this discussion. The central bank has been talking about higher capital levels, higher risk weights in Ireland, for a number of years. 

I remember being in the Oireachtas three or four years ago and talking to Michael McGrath at that point. So we’ve been endeavouring to have a better conversation. 

I do think there are some misconceptions that need to be addressed because we don’t have this machine, or model, where we’re churning out the numbers and imposing them on the banks. So banks are following the rules and regulations that are required under the capital requirements directive. And their internal models are generating their risk weights. 

SK: But they generate the internal models in order to be passed by you, right? So you can say it’s their models, but they’re effectively your models because the banks are doing what they have to do, in order for you to approve them. 

ES: Well, there are different perspectives on this. They are their models. But I take your point, they are subject to challenge by us. So they need to be of a certain standard. But even within the Irish banks, you can see there are different risk weights coming out of their own models. So that is some degree of confirmation that they are their own models. Fundamentally, when we were talking about mortgages, the loss given default, the time in default, the probability of default are important factors. We have really engaged with the EBA (European Banking Authority), and also the SSM (Single Supervisory Mechanism), in terms of transparency. I will certainly take that away and talk a little bit about whether we could do more.

Read more:

Why high interest rates suit the Central Bank just fine

After 165 years here, why can’t Ulster Bank make money in Ireland?