ING Groep NV (ING) 2020 Q4 法說會逐字稿

(technical difficulty) future financial performance and any statement not involving a historical fact. Actual results may differ materially from those projected in any forward-looking statements. A discussion of factors that may cause actual results to differ from those in any forward-looking statements is contained in our public filings, including our most recent annual report on Form 20-F filed with the United States Securities and Exchange Commission. And our earnings press release is posted on our website today. Furthermore, nothing in today's comments constitutes an offer to sell or solicitation of an offer to buy any securities.

Good afternoon, Geert. Over to you.

Thank you, operator, and good day to you all on the call. Thank you for joining us to the second credit update call. I'm here together with Mark Milders, Head of Investor Relations. And we also have Ewald Walraven, who is responsible for issuance and investments; and Paul van Slobbe of the IR team present potentially to answer any additional questions.

After the positive feedback of our call last year on August 7, we have -- on the second quarter results, we decided to repeat this credit update call on a semiannual basis. 2020 was the year of the virus. And hopefully, this year will be the year of recovery and vaccines. Here at ING, the majority of our employees have been working from home, partially or fully, for almost a year now. And we had to find new ways of working together, not only as colleagues, but certainly also with you, our valued investors. Unfortunately, the team and myself have been unable to come and see you on physical roadshows to meet. But we certainly appreciate all the virtual interactions that we had with you to keep you informed about what is going on in these unprecedented times.

In this call, we will take you through some of the highlights in terms of our strategy, our Q4 results, the asset quality as well as the capital position, the capital return and our plans for debt issuance. And at the end of the call, we will have ample time for Q&A.

So before we start, I'd like to point out that the credit update slides that accompany this call are available for download from our website for those of you that are not viewing us via the live webcast.

And over to Mark for now.

Thank you very much, Geert. Good day, everyone. Thanks for joining. And indeed, as Geert mentioned, we have it on the webcast for those people that are not on the webcast and looking at physical slides. I will not go page by page. We had a very extensive call already this morning with our Board. I will give some highlights, but I will call out the slide so it will be easier for you to follow.

So let me start on Slide 6, which highlights our current standing and our activities on a topic which is dear to our heart, ESG. We are very proud to be considered an industry leader in this space, which is not just something that we ourselves want to be. It's also commercially for us an important area and something which, especially in capital markets, we see as only growing faster in terms of its importance for us as a company. So what we basically see it -- the industry leadership on several angles also in disclosures. And the commercial angle is not only to be seen as green, but it gives commercial credibility as a bank that helps our clients on their green issues and social bond issues.

The Terra report tracks how our loan book evolves towards below 2 degrees Paris accord. And the business angle therein is that many of our customers that are in these more carbon-intensive industries also need to have investments for transformation in which we both advice and finance them how to do that. Furthermore, it improves liquidity for our issuance and for the funds that are able to invest increasingly. That's becoming an important factor. And finally, you also see it back in our remuneration and compliance practices, with a clear link of sustainability to performance.

Moving on to Slide 7. Especially this year, wherein people had to stay at home, where we had travel restrictions, but you still needed to do your finances, our digital capabilities are a clear advantage. This is where the real capabilities of bank apps got tested, whether it was more than just being able to check your bank balance or whether you are actually able to transact. And as you can clearly see from our data, we see an acceleration, and our digital mobile-first customer proposition has really benefited us in 2020. So mobile-only customers that only contact us via a phone or mobile device has increased. The mobile interactions have grown very, very strongly. And also mobile sales, which in the end is, of course, the proof is in the eating, but you are also -- we're also able to sell via our mobile device without anyone having to go into a branch.

The other highlight I want to point out is in the bottom right corner, which was the investment accounts. This is an example from Germany, where the ease with which on a mobile app you can open an investment account has led to 326,000 new accounts opened, of which 20% were new to bank, an increase of 25% of assets under management. And increasingly, the trades are not being done by phone or web, but actually directly in the app. And that is also underpinned that, in a period of a crisis where normally in behavior, you see that people stick with their main bank and remain with their bank that they were before something happens. We were actually quite pleased to see that we could welcome another 578,000 new primary customers.

Moving to Slide 8. We -- across 2020, which you can clearly see in the graph on the top left is the complete change in behavior brought on by the 2020 consequences of the COVID-19 pandemic and the restrictive measures that have been put in place. So as you can see, in the past, we -- our loan growth exceeded our deposit growth. And in the past, we were able to print 5% growth rates annually. What you see now in 2020, as clearly as investment and businesses, with the growing uncertainty delay investment decisions, but also have a lower requirement for working capital need, you see that lending has shrunken. And you also see, on the opposite side, on the liability or deposit side, you see that people in lockdown spent a lot less holidays, less bars and restaurants, things we all want to do again, but clearly, have led to an inflow on our balance sheet.

Now if you look at NII, the other things that have been a clear headwind, but have already been a clear headwind for 4 years is the negative rate environment. And we've been able to manage those quite well in the past years. Clearly, COVID and the lockdown restrictions have made some of the levers that we utilize have made them less effective. So besides loan growth, you see that, as one of our strongest lever, that has not worked and what you also see is that in non-euro countries, you have seen core rate reductions, which have also affected our NII.

Now the swap curve, at which we replicate our liabilities, grew -- went even lower in 2020 and where possible even flatter, which clearly puts the pressure on our replicating income for liabilities and that pressure remains. So FX -- sorry, I should mention, FX also had quite some effects this -- not only the quarter but in the year as well with the volatility also brought on by the pandemic.

Looking forward, more positive, as Steven, our CEO, also said this morning. We do see that with vaccination and everything, we do expect that loan growth likely to return in the second half of '21. We're already seeing a good pipeline in both Asia and the U.S. Furthermore, to counter these effects, we've introduced negative charging in several countries, Netherlands, Belgium, in Germany for new customers and a custodian fee in Spain. The tiering helps mitigate some of the pressure. And finally, more in our own control, we are maintaining very strict margin and pricing discipline, which is also clearly observed in the quarter's results, where the margins on most of our product actually went slightly up.

I move to Slide 9 on fees. Fees were arguably one of the highlights, where you see that our potential and the optionality that we have to increase or introduce fees, coming from a very much low fee, no fee history in our direct bank model is bearing fruit, where pre-COVID, we already increased some payment packages, namely in the Netherlands and in Belgium. We also introduced fees during this year on products. And we were quite successful in assisting our customers to move from savings into investment products. There, we saw a strong increase, particularly in Germany, as I highlighted earlier.

Now and these introduction of fees and this -- compensated actually for the negative effects of the pandemic on fees where, as you can understand, international payment transactions because of travel restrictions were severely depressed, but also quite depressed and at a lower level than last year, but recovering towards the end of the year, where domestic payments as people in lockdowns spend less. So that decline was more than compensated, and we are quite pleased with the result we can show today.

Moving on to Slide 10. As Steven and Tanate, our CFO, have often said, the focus is on cost. And we've seen a pressure on our top line, which means that we need to work on our cost. And Steven has called it the nose of the plane has to come down. Now as you can see here, it has. If you look at -- if you compare full year 2019 to full year 2020, the costs were actually flat, corrected for incidentals. But also quarter-on-quarter, if you correct for the incidentals, which are actually redundancy provisions and some impairments, we also show a flat cost base. If I dig one layer deeper, wholesale banking has already been showing 4 quarters in a row a cost reduction so has market leaders. And we continue, as Steven said, to take incremental steps and we'll announce them as they come. But we want execution certainty, but it's clearly the focus and it has been paying off.

Taking you, please, to Slide 11 on the risk costs. Clearly, 2020 saw the first full implementation of the IFRS 9 methodology, with the staging and also with the pull-forward effect that you have from lifetime expected loss. So the total risk costs over 2020 were EUR 2.7 billion, as you can see on the slide, which was 43 basis points over our average loan book, which I'd like to point out is significantly lower than other Eurozone peers. Of that EUR 2.7 billion, 30% is roughly in stage 1 and 2.

To highlight a couple of things. Of course, we also granted payment holidays in the beginning of the crisis to support our customers in these times of uncertainty, which was roughly EUR 19.4 billion, which only is 2.6% of our loan book, so fairly limited. As we did not get that many requests, just to point out that in most of our core markets, the government support actually went direct via fiscal measures rather than via guaranteed lending through the banks, as you can see in some other countries in Europe.

Now most of these payment holidays have expired. 93% has expired and with very few requests for extensions and no material observable deterioration in the risk or credit profile of those clients exiting the payment holidays.

So 2021 risk costs, we expect them to move close to our -- through the cycle average, which is 25 basis points over our average loan book. This is something that Steven highlighted and also our new CRO, Ljiljana Cortan, referred to is the quality of our loan book. It's something we often say and it's -- sometimes, it's waved aside from, yes, yes, let's see until the next cycle.

But I think we've got enough data points over a long period of time to show that it's not that we never make mistakes, but our risk framework with all its floors and caps on geographies, on sectors, on individual exposures as well as the diversification of our loan book even helps us maintain a healthy book even during a global pandemic. Our book is senior almost exclusively. A large part of that is secured lending. And also, I might add that 50% of our book is mortgages in strong economies.

I would like to stop here. And Geert, I hand back to you for the capital and funding part. And I'll be available for questions later. Thank you. Geert?

Thanks, Mark. I'd like to take it to Slide #23 and tell you something about the risk-weighted assets development during the last quarter. As you can see, the RWAs decreased by EUR 6 billion to just above EUR 306 billion in total. And this decrease was mainly caused by a reduction of credit risk-weighted assets, which were down by EUR 4.5 billion, in line with the knock-on effect of COVID, driven by lower volumes and a shorter duration of the wholesale loan book and a better LGD profile.

I'm also happy to say that by now, we have absorbed the biggest impact -- or let's say, we have largely absorbed the full impact of the TRIM exercise with the latest addition of EUR 2.5 billion that was related to the segment financial institutions and trade and commodity finance. And the outlook for further regulatory inflation, as we look at it today, we expect to be roughly 50 basis points. So this should give some more stability from a regulatory inflation volatility on our CET1 ratio going forward.

Please turn to Slide 24, where we'll show you the evolution of our CET1 ratio. That now has improved to 15.5%, which is 300 basis points above our ambition level of around 12.5%. And the decrease in RWA, that was partly offset by a change in the CET1 capital that decreased by EUR 0.5 billion. The current ratio implies a surplus of roughly 500 basis points above MDA, and like I said before, 300 above our ambition level. But it's important for you to note that this ratio currently excludes the almost EUR 3.3 billion of profits that were reserved outside of regulatory capital. Implicitly, that reflects an additional 1% of CET1 ratio. But we're happy to have it reserved outside of the reg cap.

And we're also happy with our current AT1 and Tier 2 ratios, where we make full use, full benefit from the relief offered by Article 104a of the Capital Requirement Directive. And as long as we have excess CET1, we'll manage those ratios close to these levels.

The next slide will be Slide 25, which shows our MDA levels. And most notable change this quarter was the reduction of the systemic risk buffer by the Dutch Central Bank from 2.5% to 0% and the increase at the same time of the O-SII buffer from 2% to 2.5%. Given the implementation of CRD V in national laws, these changes have no effect on the combined buffer requirement.

Let me take you to Slide 26 now, which is an interesting slide on our capital distribution plans for this year and beyond. As we have announced last quarter, we moved from a progressive dividend methodology to a 50% payout ratio of net resilient profit and we have also adjusted our CET1 ambition to around 12.5%. And in line with that new distribution plan, we have reserved EUR 1.5 billion over 2020, which is now added to the EUR 1.8 billion originally reserved for the final dividend of 2019. And that brings you then to a total just south of EUR 3.3 billion.

And obviously, we align with the current ECB recommendations where we will pay EUR 0.12 per share, equaling EUR 461 million on the basis of the 15% full year 2020 adjusted net profit as it is defined by the ECB. We do intend to distribute the remaining amount that was -- that is reserved after September 30 of this year, but obviously, again, subject to prevailing ECB recommendations and subject to the relevant approvals that might need to be obtained. Since the current ECB recommendations also prevent us from paying an interim dividend over 2021, today, we also announced that this will be delayed until after September 30 of this year.

Now as mentioned before, 300 basis points CET1 excess over and above our ambition level, there must be room for further distribution of capital over the coming years, where we intend to manage towards our 12.5% ambition level.

Next topic is on Slide 30, which will give you some guidance on our debt issuance plan. We meet our TLAC requirements on an RWA basis with 28.2%, but MREL will remain the most constraining factor for ING. Our current nonbinding requirement is still of -- so leverage base 10.45%, and this was based on the balance sheet of year-end 2019.

And we have received a phased-in period until Jan 1, 2024, with a linear intermediary target by January 1, 2022. And I have seen the draft -- the new draft requirements under the new MREL policy based on BRRD2. But those requirements are still in consultation, and we do expect to receive a final letter probably in Q2 this year, but certainly before second half starts.

Now being liability rates driven by the EUR 40 billion plus inflow last year of clients' liabilities, our issuance will be fully driven by meeting MREL requirements, the issuance out of the HoldCo. And based on our planning, both capital planning, MREL planning with also our business planning, we expect to issue EUR 6 billion to EUR 8 billion of HoldCo Senior for the remainder of this year. And this comes on top of our EUR 1.5 billion Euro loan nonco 8 that we issued early in January of this year.

As mentioned before, we're happy with the AT1 and Tier 2 ratios. And the driver for further actions there is clearly balance sheet developments, RWA growth and any need arising from potential calls or redemptions.

Given our participation in TLTRO III and in combination with the increase of our clients' liabilities, we do not foresee any long-term OpCo Senior issuance this year. We might jump into interesting covered bond issuance durations by some of our subs, potentially in green format, but it will be opportunistically driven. And surely, again, the development of the balance sheet for this year, we closely monitor.

The focus for the team will always be to meet capital and liquidity ratios. But we also clearly take a careful look at our net interest paid on the back of al. this issuance. So that plays a very important role as well.

Slide 32 touches upon our green bond framework and our sustainabilities. As you might recall, in 2015, ING was one of the first G-ship issuers of green bonds. And also, in recent history, we were the largest issuer of a multiple tranche green bond in the U.S. Over the years, we have further developed and improved our green bond framework, aligning it with the regulatory developments. And we are prepared if and when it will be endorsed by the EU parliament to adhere to the EU green taxonomy. We intend to remain active in this senior unsecured green format as well as in green covered bonds. On top of that, we continue to evaluate the market for other ESG-driven issuance opportunities such as social bonds or maybe even a Green Tier 2.

Having said all this, and before I stop talking now, there should be some time for Q&As on the back of this presentation. So I hand back to the operator and see if you have any questions.

(Operator Instructions) The first question is from Lee Street, Citigroup.

I have 2 for you, please. Firstly, when you talk about excess capital, it always seems to be in the view of returning it to shareholders and buybacks today and also on the call this morning. From that, should I assume that it's very, very unlikely that we would see ING to take any form of significant M&A transaction within your preexisting markets and that capital really is just earmarked for return to shareholders?

And secondly, just -- we obviously had the EBA opinion on the legacy instruments last year and the discussion around infection risk. I know thing you're asked about on your last call as well. Having seen that now with the BRRD being French bankings law, can you just talk us through how you see the existence of your older legacy Tier 1 securities, how that impacts across the cap structure? That will be my 2 questions.

Yes. Lee, thanks. I'll take the first one. Steven got the same question this morning. Basically, we see a couple of sources for use of our capital. One is to build it for regulatory requirements. We believe we have about 50 basis points left to go of regulatory RWA that we need to absorb until implementation of Basel IV. So that's one.

The other one is to fund loan growth. And then you have, indeed, as you pointed out, distribution and M&A. But as Steven clearly said, we're a digital data-driven bank. We focus on the customer experience. So if you think about M&A, we'd like to say that we're more looking for car parts than cars. A favorite comment also, of course, dimmer loans in the past. And that basically is, if we are in a position to buy technology which helps us become more efficient or increase our customer experience, we will have a look at it. Similarly, you could think that we might look at portfolios that augment our current strategy. But the focus is on organic growth.

Then, Lee, for your second question, I ask Ewald to provide you with an answer.

Yes. Thank you, Geert. Lee, thank you for your question. We've been debating or in discussion on the legacy -- on the remaining legacy grandfathered hybrids in -- that we still have outstanding. We're talking here about the EUR 1 billion denominated. Yes. We've taken notice essentially of the EBA opinion, which was published in October of last year on the future treatment of grandfathered instruments.

So in essence, they rank pari passu with our CRD V compliant instruments and, therefore, could indeed create what in the industry is also flagged like a potential infection risk. Now the element, which is at this stage unclear or needs to be clarified, if you will, is that we are still awaiting the implementation of BRRD2 in the Dutch and solvency law to see what the, let's say, local approach of that will be.

So at this stage, we don't have the full granularity. The Dutch government opened a consultation phase, which has ended by the end of January. And hopefully, in the coming weeks or months, we will have more insight in, let's say, the Dutch approach to implementing the insolvency law. So that's also when at least we can make further analysis internally, engage with our legal councils and clearly also with the...

With our regulators. In addition, these are instruments that are issued to retail investors. So that could -- they could be deemed as an impediment to resolution by the SRB. But as I mentioned earlier, this is, in essence, too early to tell, and we await the outcome of the consultation.

Okay. That's really helpful. Just one quick follow-up on that. In terms of the consultation that was set out and has now closed, did that -- was anything proposed to resolve the infection risk in the statutory law or not as a part of that consultation?

Lee, we participated in the consultation, but we don't comment on what we've said.

The next question is from Mr. Daniel David, Autonomous Research.

I think you've covered most of my infection risk questions just previously. So I'll just touch upon green bonds, if that's all right. Noting that the progress in your commitment to the space, is there a percentage kind of green bonds versus familiar issues that you see? So -- and could you get to 100% green issuance in the senior layer? Also, are you seeing a material financial benefit from issuing in green format? And more broadly, would you consider issuing an ESG format linked to maybe carbon targets or ESG ratings?

Thanks, Daniel. Let me take the first part, and then I hand over the carbon linked question to Ewald. Obviously, our green bond issuance is fully driven by our green bond -- sorry, by our green asset origination. So this goes in tandem. The more green assets we originate, the more availability we will have to issue green bonds. There is not, in that sense, a percentage target of where we want to be 5 or 10 years from now. There is a target of what we want to issue in green this year, but it can be green covered, it can be green senior and it can even be a green Tier 2.

On -- in terms of pricing, we do recognize, because of the lack of supply of green products, there is such a thing like a greenium. And the greenium that we have observed is between 0 and 5 basis points more attractive than a similar instrument without the green label. So that is what we see there in terms of pricing.

Ewald, can you take the carbon-related question?

Yes. Thank you, Daniel, for your question. In essence, we've now focused on CO2 reduction in our green bond impact reporting, which is also calculated by external parties and which is clearly published on our website. I would not say that we have direct links or targets actually in that sense. But clearly, it's an element we take in consideration.

The next question is from Mr. Robert Smalley, UBS Fixed Income.

Two things, both related to loans. First, there's an idea that the longer that we see government stimulus in -- around the world that this not only will this bridge us to a better economy, but it will actually help reduce the amount of outstanding problem debt as -- the longer that, that goes forward. Do you see that as being the case in the Netherlands and for the bank? Or do you see that differently? Is my first question.

And my second question is on loan demand. Certainly, on the consumer side, a lot has been talked about with respect to pent-up demand for purchases and paying down credit cards. Could you talk about how you see it on the corporate side? How quickly you think that loan demand will get back to pre-COVID levels?

Okay. Thanks, Robert. So on the first one, on the government support, you can dissect a little bit how it -- how the support is being given, roughly Northern Europe, Southern Europe. Governments in Northern Europe are providing the support fiscally. So that is either with support for wages or a deferral or delay or reduction of tax burdens. Whereas in Southern Europe, you see a preference for doing it through government-guaranteed lending, but then that's going via the banks.

Now what the outcome of the both directions is, they shouldn't be that dissimilar. But you also see that the borrowing in Northern Europe is just less because of that because it's just either delayed cost. You also have slightly better labor laws than in some other countries. So you have a more flexible workforce in countries like the Netherlands where you can upscale and downscale through temporary workers a little bit better than in some other countries where there are stricter labor laws.

So that's on the fringes. You could argue that your statement that it will help to reduce outstanding problem that we would subscribe to that. You must realize that I think that the whole world is still coming to grip on what business models have actually become obsolete. So there, I don't expect that there will be absence of problems. That's not at all. But the government support has been a very, very strong help in helping the European countries face this humanitarian health crisis.

On loan demand, so specifically on wholesale banking, I got from your question that the loan demand on consumer lending is clear. I mean what happens in every crisis, people defer large payments. And with the lockdowns, people will probably also think twice about buying a second car or even a first car.

On wholesale banking, good to remember how the crisis works on the business. There's 2 effects. Every management team will probably scratch themselves behind the ears whether or not to build that second factory. So you see a drop in investment -- demand for investment loans expansion.

The second thing that happens, and especially now because you have both a demand and a supply shock crisis, you see that working capital reduces dramatically. And I think that has also been the cushion for many companies. That liquidity that was trapped normally in your production cycle, that was released as people had to produce less and also had difficulty restocking because of -- especially in the first half of 2020, you had a logistical meltdown of the material goods not reaching the factories or the businesses.

Now how does it pick up? That was the second part of your question. At what speed? First thing that comes back is working capital need. Working capital need will come back as quickly as you want to ramp up your production again and as demand increases as economies open up and, hopefully, the vaccine takes the -- what we all wish for, has the effect of bringing us back to more normal. So you have -- on the demand side and, therefore, with the control of the virus spreading, you would also see a recovery of the logistical networks. So that will be our fastest recovery.

Then on investment -- demand for investment loans. Basically, that depends on the confidence and the clarity of the outlook for people that will depend where you are in the world. I would argue that ING is -- with its geographical and product diversification, as in the past, where we used to print 5% growth, we're ideally placed to capture the growth wherever it takes place geographically. We already see a pickup in our pipeline in both Asia and the Americas.

The next question is from Mr. Alexander Latter, PGIM.

Sorry to come back to this. I know you talked about it a lot today, but just for my own clarification on the kind of Basel IV and the capital walk. Your 50 basis points of capital headwinds, that's sort of pre-Basel IV. What is the guidance now, if any, for what your kind of pro forma Basel IV number is? And where you would be versus that 12.5% if we fully loaded it today?

Yes. Thanks, Alexander. Basically, the 15.5% that we report and that we mentioned as well as our 12.5% -- around 12.5% ambition are both Basel IV numbers. So the only thing that we're saying is that from the remaining TRIM exercises as well as the implementation date effect, you would expect then a 50 basis points RWA inflation still to happen. Initially, what we said already 3 years ago is that, in total, we would expect, absent from management action, a total impact of around 200 basis points of -- but that was then only Basel IV. TRIM was partially -- came later, was partially front-loading Basel IV, but some elements are not. So we actually stopped calling it Basel IV headwind and just call it regulatory RWA effects. And of that, there's 50 bps remaining, we estimate, of what we can see today. And as you know, we generate annually anywhere between 140 and 170 basis points of capital.

And then how much are you suggesting the kind of implementation day effect would be?

That's probably almost all of the 50 bps. There's a little bit of some TRIM still to follow on some smaller portfolios. But you can imagine that the 50 bps, whether or not you have at that moment in time, which is now '23 or '24, the rumor say '24, but the last number we have fixed is January 1, '23, is only then can you determine whether or not you have any binding constraints from your output floors or not.

So on the day that Basel -- some of the elements of Basel will only take effect if that comes into law and effect on the date of the implementation. On that date, you will have an effect. But on the other side, some of the management actions that we've been taking, such as rating non-rated corporate clients, which helps on the LGD, will also only have an effect once Basel trade comes into effect. So we also have some management actions to prepare for, which are only effective as of the date of implementation.

(Operator Instructions) The next question is from Mr. [Bharat Manjari], Schroders Investment Limited.

Just one. I think you mentioned in terms of your guidance for cost of risk for 2021, you're expecting it to come back to normalized levels this year, which, if I compare with all the other European banks guidances, is much more optimistic. I wonder what sort of assumptions you have made in terms of how quickly sort of vaccine is rolled out or how quickly the restrictions are lifted. And what might be the downside risk to that guidance? And also, if you could give any sort of views on how the nonperforming loans should develop from here? And do you expect them to be at certain time this year?

Thank you, (inaudible). I think you're going to have to repeat the last part of your question. I didn't quite catch that, but I'll start off. So I wasn't aware that we're that much more positive on our guidance than other banks. I think most banks are guiding towards lower risk. And that's simply also an effect of IFRS. IFRS were, in case of a macroeconomic outlook, the macro part of that model, basically, if you have a sharp deterioration demands, that you take certain actions and also pull forward some of the expected losses on base of the macro and what you generally see is that more clients -- because the outlook deters as you put on to watchlist, so that takes a Stage 2.

So this was -- what you saw happened in Q1 and Q2 last year, clearly, because visibility was that low. And as I sometimes joking to say, everybody was googling, including the macroeconomists. So as you then progress, basically what happens under IFRS, you have a forward rolling quarterly macroeconomic outlook model, and a bad quarter rolls off and a good quarter rolls back in, which leads to what you already saw in Q3 and Q4, to a macro model IFRS release in risk cost. Now in the third quarter, we countered that because we thought it was a bit counterintuitive to have releases and risk costs right at the moment that the second wave was hitting and all of Europe was going into second lockdown. That's #1.

On the other hand, we still had many clients in payment holidays, and we're not sure how they would emerge from that. So as a prudent bank, both on our -- who we put on our books, but also on how we manage risk, we then took a management overlay.

This quarter, you saw a double effect and that's not only did a bad quarter roll off out of the macroeconomic model, you also see that the macroeconomic outlook, where for which we use a consensus outlook, has improved, which was partially driven, of course, by the approval of several of the vaccines. So you have a double effect. You have out of the data pool for your macro models, a bad quarter rolls in and a recovery model quarter rolls in as well as the economic forecast of GDP growth have been revised upwards.

Now why are we so confident on then making a guidance that we believe that next year, our cost of risk should move towards the more average across the cycle level of 25 basis points? It's also IFRS, which basically has this pull forward effect. So Stage 1 is your expected loss on your -- at inception of a loan over the entire loan book. Then you have Stage 2, that is for credits on watchlist or where you have a heightened sense of concern, be it risky sectors or whatever the reason is to put it on watchlist, but still performing.

And then you have your Stage 3. We believe that we've taken a conservative and prudent approach, knowing our loan book and the behaviors of our sectors, especially in times of crisis quite well. We believe we have provided sufficiently in 2020 to cover possible loan losses that can be expected in 2021.

So just to clarify, I think I misunderstood. So you're guiding to cost of risk to normalize this year or next year?

Sorry. This year, '21.

So you're guiding to this year? Yes. Okay. Because most of the other banks are guiding to normalization next year. So that's a bit optimistic. Okay.

The other question I had was, I think -- I mean, yes, payment defaults or payment holidays, they have turned out to be much better and much sort of normalized level sort of assumptions. I guess, in my view, the risk is probably now remaining on your more sensitive sectors and how -- as you said, how the business model develops once the restrictions are lifted. Are you -- are those sectors -- again, thinking about how the bad loans should develop, are the reliance to a significant exchange on the government support schemes, is that another reason why you're thinking that the provision that you have taken in 2020 should be sufficient enough and then that should basically cover off any sort of increase in bad loans here? Because I understand the exposure for ING is probably not that high, but it is still quite a decent amount. So I'm just -- I guess, in my mind, the question is, what level of sort of government support is baked in with that assumption?

Right. And you're right. So if you look at the most formable right now in society, it's probably small, medium enterprises. Of which, frankly, we -- if you look on a relative basis, we, as ING, have a very small proportion. If you look at the analyst presentation in the Annex, we spelled out for you actually how much we have in the various segments of business lending, wholesale lending, consumer lending and also made a breakout for you for what we consider to be the more sensitive sectors, like hospitality, nonfood retail, aviation. You can think of the rest.

It's -- but if you think about those sectors, they are so small in perspective to our bank. So nonfood retail, for example, 0.4% of our loan book is EUR 3 billion. Hospitality and leisure, if I include both wholesale and business lending, is 6.1, so 0.8% of our loan book. So either by wisdom or serendipity, I'll let you choose, but we've managed to stay away from the very risky sectors that are affected by COVID.

There are no further questions at this time.

Okay. Thank you, operator. Thank you all for participating in today's call. The IR team, of course, is available for any follow-up questions. But we all certainly hope to see many of you again in person during the course of this year. That will be very positive. So I conclude by telling you to stay safe, and goodbye for now.

Ladies and gentlemen, thank you for attending. This concludes the ING Fourth Quarter Full Year 2020 Conference Call. You may now disconnect your lines. Have a nice day.