Good morning, ladies and gentlemen. Today's conference call will be hosted by UniCredit CEO, Mr. Jean Pierre Mustier, and CFO, Mr. Stefano Porro. At the end of the presentation, there will be a question and answer session. Today's conference call is being recorded. At this time, I would like to hand the call over to Mr. Jean Pierre Mustier. Sir, you may begin.
Thank you very much, and good morning and welcome to the analyst call for our third quarter results. These are unprecedented times, and the emerging second wave of the pandemic means that many of our countries face a challenging period. On behalf of all of us at UniCredit, I would like to express our sympathies for anyone impacted by COVID-19. We will continue to do everything we can to protect the health and safety of our colleagues, and we will continue to support our clients, the real economy, and our communities in the countries where we are present, j ust as during the first wave, we are committed to being part of the solution. As you will have all heard, Pier Carlo Padoan was recently co-opted to our Board of Directors as Chairman Designate. Before we start, I would like to warmly welcome Professor Padoan.
His extraordinary professional experience and deep knowledge of Europe and the regulatory environment, as well as his outstanding public service record in Italy, will be of great value to our group. I very much look forward to working with him. I would like as well to thank Mirko Bianchi for his hard work and dedication as Co-CFO. Mirko was instrumental in the Transform 2019 and Team 23, and will bring substantial experience to his new role as CEO of the group Wealth Management and Private Banking, where we aim to become a recognized leader in Europe. Stefano Porro has assumed sole responsibility for the CFO function, and will present the results with me today. Turning to our results, we saw a promised recovery in underlying net profit this quarter, up 31% quarter-on-quarter, thanks to lower loan loss provisions and higher revenues.
The pickup in revenues was driven by increased customer activity and higher fees following the easing of lockdown restrictions towards the end of the second quarter. At the same time, we maintain our strict cost discipline, leading to positive operating jaws that gave us a more than 3% point improvement in our cost-income ratio over the quarter. Our commitment to cost discipline means that we are always looking to further optimize our cost base. Following the substantial reduction delivered during Transform 2019, we committed to further growth savings of EUR 1 billion as part of Team 23 last December. Thanks to our continued cost control and the acceleration of changes in our business model, these cost savings are now expected to reach EUR 1.25 billion, some EUR 250 million better than the original Team 23 target. As a result, we expect costs to be flat between 2020 and 2021.
Our stated cost of risk guidance for 2020 is confirmed at 100 basis points to 120 basis points due to our willingness to anticipate conservatively future impacts, including increased overlays, proactive classification, and regulatory headwinds. The combination of this conservative approach and seasonality means that the stated cost of risk in the fourth quarter will be materially higher than the 53 basis points in the third quarter. In the face of the emerging COVID-19 second wave, having a solid balance sheet and a conservative approach to risk are very important. With record levels of capital, our fully loaded CET1 ratio is at 14.4%, and our fully loaded MDA buffer is at 538 basis points, o ur liquidity coverage ratio is at 183%. We are, therefore, in a strong position to continue supporting the economy while returning capital to our shareholders.
Our culture and values are a fundamental part of how we run the bank. Our commitment is and remains to do the right thing for all our stakeholders, favoring long-term and sustainable outcomes over short-term solutions. This means doing the right thing for our employees, for whom we want to be not only the best place to work, but also, as I have already said, a safe place to work. The health and safety of our colleagues and our customers will always come first. Our ongoing commitment to sustainability received further recognition this quarter. We are ranked number one globally for sustainability-linked loans by Bloomberg, and were awarded Best Social Impact Bank in Europe by Capital Finance International. Let's turn to our key financials on slide five. Stefano will take you through these in more detail later but I wanted to highlight a few key points.
Revenues for the quarter were up 4.4% compared to the second quarter and broadly in line with our performance in the first quarter. This progressive easing of lockdowns across many of our core markets had positive effects on client activity, where we were primed and ready to capitalize on. Of particular note are the recovery in investment fees of 12% quarter-on-quarter t hanks to robust AUM sales, primarily in commercial banking in Italy, we delivered a strong quarter despite seasonality, and the comparison against a particularly strong period last year. Our third quarter 2020 underlying return on tangible equity reached 5.4 percentage points, an improvement of 1.3 percentage points quarter-on-quarter, in spite of the extraordinary macroeconomic backdrop. This reflects our continued conservative approach to running the bank, our tight cost control, and our strict underwriting discipline.
The reduction in gross NPEs continued at a brisk pace, with a further reduction of EUR 1 billion over the quarter, down more than EUR 6 billion over the past six months. This reflects continuing hard work and disposals within non-core. As a result, our gross NPE ratio for the group improved further to 4.7%. If we focus on the figure for the group excluding non-core and use the EBA definition, our NPE ratio remains at 2.7%, below the average of other European banks. Our capital position strengthened further in the quarter with the fully loaded CET1 MDA buffer reaching 538 basis points. This represents an increase of 58 basis points over the quarter and is 286 basis points higher than a year ago.
This improvement is thanks to lower-risk-weighted assets, mainly resulting from lower overall loan volumes in CIB, and the increase of guaranteed loans in commercial banking in Italy.
Now it's my great pleasure to hand over to our CFO, Stefano Porro, who will take you through the details of the quarter. Stefano, all yours.
Thank you, Jean Pierre, and g ood morning, everyone. Revenues stood at EUR 4.4 billion, reflecting the recovery in client activities after the easing of lockdowns, up 4.4% quarter-on-quarter. This recovery was supported by stronger fees, up 6.4% quarter-on-quarter. Costs were lower by 1.4% quarter-on-quarter, reflecting our strict cost discipline and continued focus on further efficiency gains, which more than offset COVID-19 related expenses. I will comment on revenue and cost components in more detail in the following slides, but I would like to make three specific comments on items below the net operating profit line. System charges in the quarter increased by 36.4% year-on-year, mainly due to additional contribution to the deposit guarantee scheme in the third quarter of 2020 for two single names in Austria and in Italy.
As a result, we have raised up our full- year 2020 outlook for system charges guidance from EUR 900,000,000 million to EUR 995,000,000 m illion. Profit on investment in the quarter was affected by mark-to-market losses of EUR 126 million on our remaining stake in Yapı. This a group tax rate of 12.4% in the third quarter of 2020 was positively impacted by a tax credit recognition in Italy thanks to NPE disposal executed in the quarter, in line with the Cura Italia decree. Let's turn to slide eight. This quarter saw limited non-operating items. Underlying net profit stood at EUR 692 million. For further details, please see the annex on page 40. If we look at the distribution of underlying net profit across the group in the quarter, you can see that our diversified business model underpins our performance at group level.
In the third quarter, all business divisions were profitable, with standout contribution from CIB and CER. CIB benefits from high revenue growth in the third quarter, mainly thanks to XVA, which led to an underlying net profit of EUR 394 million, up 87.2% quarter-on-quarter. CEE's third quarter underlying net profit of EUR 22.6 million, up 2.4% quarter-on-quarter, was underpinned by lower LLPs and continued cost discipline. Let's turn to slide nine. Before looking at net interest income in detail, let me remind you our commitment to generate long-term sustainable returns. This means that we do not do volume lending nor carry trades. We will not compromise our future asset quality to boost net interest income in the short term. Focusing on the quarter, net interest income was down 3.8% quarter-on-quarter. This can be attributed to three factors. First, loan volumes.
Just under half of the reduction in the contribution from loan volumes came from the repayment of revolving credit facility within CIB, reversing some of the benefit seen in second quarter 2020, the rest of the reduction being CE. German corporate clients refinance liquidity lines in capital markets, and a rundown was in capital financing as their sales recover. Second, guarantee business. Guaranteed loans are good business opportunities given their low risk and capital consumption. Such loans, however, are not only written at lower spread than normal loans of an equivalent maturity, they also substituted higher-yielding short-term loans in the quarter. This mixed effect was a particular important driver for commercial banking Italy. Third, three-month Euribor, t his fell almost continuously throughout the quarter, reaching historic lows by the end.
Its average value dropped by 17 basis points compared to a prior quarter, impacting all floating rate euro asset and liabilities. While NII continues to feel the impact from reduction in base rates, the net impact on group Net Interest Income in this quarter was limited. The successful repricing of deposit offset the lower contribution from both customer loan volumes and rates. With further deposit repricing becoming more difficult, Net Interest Income in CE is not expected to pass until second quarter 2021. The quarter also saw an initial contribution from TLTRO III of +EUR 18 million. As a reminder, we have accounted for the net benefit in a conservative manner, assuming our investment at the ECB deposit facility rate and amortizing the benefits across the full year maturity.
Looking forward, assuming three months EURIBOR remains at current levels, we expect net interest Income for the group in the fourth quarter 2020 to be broadly in line with the third quarter 2020. For full- year 2021, we confirm our guidance of EUR 9.5 billion. Let's turn to slide 10. While we normally discuss fees year-on-year, I want to focus my remarks today on a quarter-on-quarter basis to better capture the underlying dynamics. Fees were up 6.4% quarter-on-quarter as client activity increased despite the third quarter being seasonally weak. Please remember that if you do make year-on-year comparison, third quarter 2019 was an exceptionally strong quarter, characterized by an unusual lack of seasonality. Once again, we have seen a varied increase in commercial activity in terms of geographies and fee categories. In part, this reflected where each market was in terms of lockdown cycle.
While our Western European market had exited lockdown, in CE, some countries only entered lockdown during the third quarter. The annex on page 34 provide a detailed managerial breakdown of monthly fee trends by geography and product. I would therefore like to limit my remarks to the following. As Jean Pierre already said, investment fees deliver a strong performance, up 12% quarter-on-quarter. This reflected robust growth under management sales as client activity shifted away from assets under custody products, especially in commercial banking Italy, driving a material increase in upfront fees. We expect the recovery investment fees to continue into fourth quarter 2020. Financing fees were impacted primarily by lower activity in corporate investment banking. While that capital market has had an outstanding year, it could not match the activity of the prior quarter due to the usual seasonal slowdown in August, which also impacted structural finance.
Financing fees are expected to enjoy the usual seasonal pick-up in fourth quarter. Transactional fees recovered in third quarter 2020 as expected. GDP-sensitive transactional fees, such as cards and payment services, rose. Overall, October saw a continued positive performance for fees at the group level. For full- year 2021, we confirm our guidance of EUR 6.4 billion. Let's turn to slide 11. Trading income in the third quarter was up 27.7% quarter-over-quarter at EUR 455 million, thanks to a significant increase in XVA. Client-driven trading income excluding XVA stood at EUR 246 million in the third quarter, down 33.6% quarter-over-quarter. This was mainly due to the equities and commodities business, where there was less structuring of certificates as client activity shifted more towards assets under management products, a move which, as previously mentioned, has supported our investment fees.
There was also an impact from lower flows in fixed income and currency due to August seasonality. Non-client driven trading income was down 31% quarter-over-quarter, mainly due to fair value adjustments. Trading income excluding XVA is normalizing in line with our quarterly guidance of around EUR 350 million on average. For full year 2021, we confirm our guidance for trading income of EUR 1.4 billion. The lower contribution from dividends year-over-year was driven by the strategic disposals of stakes in Yapı and Nuju Bank over the last 12 months. However, the trend quarter-over-quarter showed an increase, mainly thanks to a recovery in the profitability of financial investment in Austria. For Yapı, intragroup funding fell to EUR 1.1 billion in the quarter, 58% below the level of quarter 2018, and thus delivering on our target set back then of halving our intragroup exposure.
Let's turn to slide 12. On cost, I would first like to remind you of the scale of cost-cutting done during Transform 19 and what is underway in Team 23. In Transform 19, we decreased FTEs by over 14,000 and branches by more than 900. This led to net cost savings of more than EUR 2 billion, materially beating the original target. In Team 23, we plan to decrease total FTEs by a further 8,000 and branches by an additional 500. As we explained at the Capital Market Day 2019, this would generate gross cost savings of around EUR 1 billion, equivalent to 12% of our 2018 Western Europe cost base. Between the two plans, we'll have delivered a 24% reduction in FTEs and a 38% reduction in branches.
Yet, in addition to these sizable planned reductions, we have reduced costs both year-over-year and quarter-over-quarter, and now expect to beat our original Team 23 target for full- year 2020 by more than EUR 250 million. As a result, we are targeting updated gross cost savings of EUR 1.25 billion, 25% better than our initial plan. Our continued cost discipline allowed us to fully offset COVID-19 related costs in quarter and full- year 2020. Total COVID-19 related costs amounted to EUR 18 million in the quarter and EUR 88 million in the nine months. For the full year, such costs will be around EUR 100 million and will be fully absorbed by savings made elsewhere. These include a significant reduction in variable compensation equal to around EUR 100 million less compared to last year.
Please remember that we expect higher costs in fourth quarter 2020 due to seasonality, both in terms of HR and non-HR. For the full- year 2020 overall, but also for full- year 2021, we confirm our guidance of flat costs related to full- year 2019. Let's turn to slide 13. First, let me remind you of our approach to provisioning that we introduced during 2020. The aim is to proactively capture the future cost of default in the loan portfolio and properly reflect the forward-looking economic component of COVID-19. Loan loss provisions, therefore, include overlays as well as specific provisions and regulatory headwinds. We also explain that since part of the loan portfolio is likely to be under moratoria for a large part of full- year 2020, specific LLP will likely be lower than it would otherwise have been.
This has been the case with the moratoria on Italian SME loans, having been extended January next year. For further details of our approach, please see page 49 in the annex. We also provide further disclosures on the staging of our loan portfolio on page 48. Let me now turn to our performance in the first nine months and explain how we expect provisioning to evolve for the rest of this year and into next. Our cost of risk in nine months 2020 stood at 81 basis points. Within this, 33 basis points were accounted for by specific LLPs for loans that were in stage 3 or moved to stage 3 during the first nine months. This is better than what we have reported in the past. It is aligned with the cost of risk guidance that we gave at last December's Capital Market Day.
Despite a third quarter 20 cost of risk of 53 basis points, we expect a significant quarter-on-quarter increase in LLPs in fourth quarter 2020, driven by a number of factors. As well as the usual seasonality, these include the anticipation of future impacts from risk overlays, proactive classification, and regulatory headwinds. The latter, mainly connected to the new definition of default, are estimated at around 10 basis points, forming the bulk of the expected regulatory headwind LLPs in full- year 2020. Our stated cost of risk guidance for full- year 2020 is confirmed at 100 basis points to 120 basis points, with lower specific provision of 40 basis points to 50 basis points and higher 50 basis points to 60 basis point overlay, as well as 10 basis points of regulatory headwinds.
The higher level of overlay provisions we are taking this year underlines the credibility of our full- year 2021 cost of risk guidance. As some of these overlay provisions will be used next year when higher defaults should materialize. Full- year 2021 stated cost of risk guidance is confirmed at the bottom end of the 70 basis points to 90 basis point range, with the underlying cost of risk close to 60 basis points. Let's turn to the balance sheet, starting with asset quality on slide 15. In the third quarter, our gross NPE ratio for the group excluding non-core, was stable at 3.5%. Using the EBA's definition, the group NPE ratio excluding non-core at 2.7% continues to be better than the average of other European banks. The reduction in the coverage ratio of 0.8 percentage points in the quarter compared to second quarter 2020, was driven by three factors.
First, disposals of bad loans. Excluding such disposal, the coverage ratio of bad loans would have been higher than in second quarter 20, above 71%. Second, the classification into UTP of a single main government-guaranteed loan, where LLPs apply only to the uncovered part, and the coverage ratio on the full loan appears below the portfolio average. Excluding such impact, the coverage ratio of UTPs would have been well above second quarter 2020, at over 50%, and t hird, an increase in likely to pay loans from precaution classification, leading to a different mix effect. Let's turn to slide 16. We continue to work hard on the non-core rundown, and the process remains well on track. We confirm our target of gross NPE below 4.3 billion by year-end, as well as the full rundown of the non-core in 2021.
In the third quarter, we carried out a successful transaction with illimity, selling an Italian semi-secure non-performing loan portfolio, a testament to our ability to execute deals even in the current economic environment. As a result, gross NPEs in the non-core were down EUR 1.1 billion in the quarter to EUR 5.9 billion. The CMD 2019 P&L guidance is confirmed. We also confirm that the overall non-core portfolio is provision to sell. The remaining financial impact of non-core on group performance for full- year 2020 and full- year 2021 is minimal. The net economic risk embedded in non-core rundown is close to zero. Let's turn to slide 17. Our CET1 capital, fully loaded, is at 538 basis points buffer over our MDA level. To put this into context, our MDA buffer is now significantly larger than our current market cap.
The 58 basis point increase in the MDA buffer over the quarter was thanks to a reduction in risk-weighted assets of more than EUR 10 billion, repeating the pattern we saw in the first quarter. Lower risk-weighted assets were driven mainly by lower overall loan volumes in CIB and changes in the risk density of our overall loan book. The latter was mainly driven by changes in the mix in CIB, an increase in guaranteed loans in commercial banking in Italy. In line with the economic recovery anticipated in 2021, we expect loan growth of at least EUR 10 billion for the next year. As a result of this and further procyclicality, a large part of the positive risk-weighted asset dynamics this quarter should reverse next year. Please note that the procyclicality effects in the quarter came mainly from PD rating migration.
We continue to accrue a cash dividend in the third quarter equivalent to 30% of underlying net profit. Note that the planned share buyback, which makes up the balance of the 50% capital distribution in full- year 2020, will only be deducted from capital once we have the approval from the regulator and the AGM. The change in the regulatory treatment of software asset introduced in the so-called CRR quick fix will take effect in four quarters. The benefit is expected to be around the mid-teens in terms of basis point. Regulatory headwinds are expected to be less than 0.2 percentage points in full- year 2020, net of the changed software treatment, and less than 1.4 percentage points in full- year 2021. This includes both TRIM and PD rating migration.
The latter totals 0.7 percentage points in full- year 2020 and full- year 2021 combined, of which 0.40 percentage points were taken in nine months 2020. Please remember that this rating migration will revert over time as GDP recovers through the cycle. Looking forward, we expect our MDA buffer in full- year 2020 and full- year 2021 to be well above 300 basis points, which is above our target range of 200 basis points to 250 basis points. We remain committed to gradually returning excess capital once regulators allow. These will be the days on the sustained excess capital over our target MDA buffer. Let's turn to slide 18. In line with the strong increase of our CET1 MDA buffer, our TLAC MGA buffer has increased to 648 basis points, well above our target range. The pre-funding carried out for our 2021 subordinated TLAC needs contributed positively as well.
Despite the tough environment, we executed EUR 1.25 billion of senior non-preferred in July and another $1 billion in September, raising money at attractive rates. For the rest of the year, we are not planning any further public issuance other than MREL-eligible products. Let's turn to slide 19. Finally, a quick look at our tangible equity, which was stable quarter-on-quarter at EUR 50.9 billion. Retained earnings were offset by a decline in the FX valuation reserves of EUR 4.5 billion, mainly due to depreciation of the Russian ruble, Czech koruna, and Turkish lira. Let's turn to slide 21. Jean Pierre, back to you.
Thank you, Stefano, and well done for your first quarter quarterly presentation. As I said at the start, this quarter was characterized by a pronounced recovery in underlying net profit, driven by lower loan loss provisions, as well as a rebound in revenues as customer activity picked up. The robust performance of investment fees was noteworthy. Thanks to our strong balance sheet and commitment of our team members, we are very well-placed to keep supporting our clients, whatever the environment. As always, we will continue to run and manage the bank in a conservative and disciplined way and prepare for all eventualities. As always, the health and safety of our employees and customers comes first.
While mindful of the potential impact of the second COVID-19 wave, we confirm our underlying net profit target of above EUR 800 million for financial year 2020 and of EUR 3 billion to EUR 3.5 billion for financial year 2025, as well as our 2023 underlying return on tangible equity of 8%. We have also improved our gross cost-saving target for the plan by 25% to EUR 1.25 billion. We also confirm our stated financial year 2020 cost-of-risk guidance of 100 basis points to 120 basis points, as well as our stated financial year 2021 guidance at the bottom end of the 70 basis points to 90 basis point range. Both years include around 10 basis points of regulatory headwinds. This is why we confirm our 2021 underlying cost of risk at close to 60 basis points.
Remember, these regulatory headwinds need to be deducted when calculating the underlying net profit that is the basis for our capital distribution. Our balance sheet remains very strong, with our fully loaded CET1 MDA buffer at 538 basis points. We expect this buffer to remain well above 300 basis points in both FY 2020 and FY 2021, and thus, above our target range of 200 basis points to 250 basis points. We remain fully committed to reinstating our Team 23 distribution policy, a cornerstone of the plan. Subject to the regulatory green light, we expect to resume distribution from calendar year 2021 onwards. Our policy remains a combination of an ordinary distribution of 50% of underlying net profit and, in addition, an extraordinary distribution of excess capital. These distributions will be a mixture of cash dividends and share buyback, acknowledging the divergent wishes of our shareholders.
For the return of excess capital, share buybacks are the preferred option as they make the most financial sense at current share price levels. We have also been very clear on what we mean by excess capital. This is the sustained excess over our 200 basis points to 250 basis points CET1 MDA buffer. As such, we will pay careful attention to all the expected regulatory headwinds, and we will update this at our Capital Market Day to be held in the first half of 2021, the date to be confirmed. We want to draw the lesson of the pandemic, ensure the concrete progress of the acceleration towards more remote banking. It also means any excess capital will be returned gradually. However, while maintaining a very strong balance sheet is important, we will not keep excess capital for the sake of it.
The efficient allocation of capital is at the heart of how we run the bank. Before we finish, let me extend my sincere thanks and appreciation to all UniCredit team members whose commitment, resilience, and continuing hard work in this unique and continually developing situation has allowed UniCredit to prosper and to do the right thing for all our stakeholders. One final housekeeping item before taking your question is to note that our virtual team this quarter will be complemented with that of Lorraine, our Head of Finance and Control. We very much look forward to seeing you all and continuing our dialogue. Now, Stefano, the rest of the team and I are ready to take your questions. If you would be pleased to be so kind as to limit your questions to a maximum of two each. Many thanks. Operator?
Thank you, Mr. Mustier. We will now begin the question and answer session. Anyone who wish to ask question may join through star and one on their touchtone telephone. To remove yourself from the question queue, please press star and two. Please pickup the receiver when asking questions. To ensure better audio quality, please make sure you are not on loud speaker. Anyone who has a question, may press star and one at this time. The first question is from Antonio Reale with Morgan Stanley. Please go ahead.
Good morning, everyone, and thank you for taking my questions, and b efore that, I also wanted to wish my best to Professor Padoan for his new role. I'm sure and confident he will be a great chairman for the bank. Two questions. First one on net interest income. I understand you're confirming the guidance of EUR 9.5 billion next year. However, some of the headwinds you mentioned seem likely to stay. I'm thinking about Euribor, which has now hit new lows in Q4, and Central and Eastern Europe activity, which was probably affected more than any other region by the second wave. How do you expect to cover for the difference? If I annualize your Q3, Q4, you are at EUR 9.2 billion versus the EUR 9.5 billion . That's my first question. My second question is on dividends. You've always been open and transparent on regulatory matters.
UniCredit has an MDA buffer now of more than double your target. That leaves, obviously, a significant excess. You've talked about it. Some of the excess, if I look at it today, it's EUR 10 billion almost, will cover for regulatory headwinds, as you said, but some of it will go to shareholders gradually. If I understand correctly, I think you've said already in 2021, of course, if ECB allows it. That's on top of the 50% ordinary dividend. Now I appreciate it's all very fluid, but can you help us understand what you think are the key criteria for the ECB, to assess the bank's distribution capacity and how much of that excess you think you could, you know, realistically pay in 2021? That's my second question. If I may, just very quickly on a clarification on the sub-holding project.
If I understood correctly, the project came up first as part of your Capital Market Day last year, which was very consistent with your de-risking efforts at the bank. The idea of the sub-holding was not much to see improvements on funding costs, correct me if I'm wrong, but rather to sort of limit the capital volatility and reduce the tail risk on the Italian sovereign. This was also in line with the target to reduce the BTP. Since then, things have changed quite a lot. The ECB has a strong commitment to QE. It could be extended in December. The sovereign-bank nexus is probably stronger than it's ever been, with all the government support measures in place, so m aybe just to help understand what would be the rationale of setting a sub-holding company in this environment. Thank you.
Thank you very much. Well, just let me take your questions one by one. I will let Stefano, if necessary, comment with more detail. On the NII side, we confirm our guidance for EUR 9.5 billion in 2021. The main reason is the same across all our main markets, which is the strong rebound of the economies and the resulting growth in loan volume. We have said that during the presentation that we expect a EUR 10 billion increase of loan volume, which will be supporting more fixed investment by your client next year. Every crisis have one thing in common, that they all end, actually. Our economies are forecasting a 5% rebound in the European economies next year. This will mean more borrowing, more fixed investment, and we will be supporting our clients for that.
That's important because the higher loan volume will be, of course, one of the drivers in terms of improvement of the NII versus the trend that you can see in Q2 and Q3, basically. Second thing is that we expect base rate to stay at the current level. On the CEE side, we expect a rate stabilization on the loan side in the course of the year, and we will see a still positive, even a diminishing contribution from deposit repricing. We had a very strong deposit repricing this quarter, as you have seen. As a result, on the C side, the NII will trough, you know, in the end of the first quarter, second quarter 2021, and then we expect the NII for the C side in FY 2021 to be up year- on- year.
In Western Europe, we have a higher loan volume, as I said, which will compensate the tighter client rate and because of the increased government guarantee loan. We will, as well, as the economy rebounds, normalize the underwriting guidelines for the loans to individuals, mostly consumer and mortgage. We have taken so far a conservative risk approach, which is important when we have such a sharp fall in the GDP. We are ready as soon as the economy turns to actually increase our loan volume, as I said, and you know, resume a more normal underwriting for consumers mortgage. All these reasons mean that we expect our NII for 2021 to be very equivalent to what we had in 2020, around EUR 9.5 billion. As far as the dividend is concerned, we have, as we said, an excess capital.
This excess capital allows us on one side to finance the economy. Stefano mentioned that the improvement we've seen in the CET1 ratio this quarter of 58 basis points is mostly due to the fact that we have the clients paying back the revolver, their revolving facility, and we will have these benefits being returned next year as we have increased loan volume, for instance. You have to look at the CET1 buffer in light of risk-weighted asset inflation, and we should have increase of loans, as I just said. In terms of regulatory headwinds, we have told the market that we should have next year less than 140 basis points of regulatory headwind. We have had a translation of regulatory headwind from 2020 to 2021.
We are planning to have 100 basis point combined regulatory headwind between 2020 and 2021 before we had the rating migration, before the crisis. These regulatory headwinds transfer to 2021. On top of that, there's the rating migration of 70 basis points, as mentioned by Stefano, between 2020 and 2021, 45 basis points taken in 2020 and the balance in 2021. This rating migration will reverse over time when the rating improves, basically. We have a rating migration which we will take as well in 2021, and regulatory headwinds of less than 140 basis points, including rating migration, regulatory headwind, and TRIM next year. Beyond that, we communicated at the Capital Market Day 2019, the different rating migration.
When we take into account the projection of the regulatory headwind of the risk-weighted asset inflation because of higher loans, we can look at the excess capital we have above our 200 basis points to 250 basis points basically target. We intend over time to return all our excess capital above our target to our shareholders. We will update all this projection at our Capital Market Day, we have substantial excess capital when you look at it over time in the way I just described. The ECB will look probably, I cannot speak on behalf of the ECB, when we apply, for instance, for share buyback, the ECB looks at the projection of capital over time to define what is the excess capital above the MDA.
We expect that the ECB will apply exactly the same policy, basically, and we will apply for share buyback next year, not only coming from our regulatory dividend payment from the net income, but also for our excess capital. We expect the ECB to only allow banks to gradually pay their excess capital next year, which is why we said that we will pay our excess capital gradually next year and over the coming years, basically. I think what is important is to look at the excess buffer above our 200 basis points to 250 basis points over time or duration of the plan in being. Sorry for this long explanation, but I think it's important for everybody to understand the way we look at our excess capital projection.
As far as the international holding is concerned, we announced last year that we were looking at the creation of an international holding in order to see if we could improve our cost of funding and/or our MREL and TLAC constraints. With the ECB extended quantitative easing and the tighter sovereign spread, this phase, this project, which has always been a project, is still a project, and b ecause of the macro environment and the ECB action, we don't intend to move on the project, because there's no reason to do it with the credit and sovereign spread being very tight. Thank you. Next question, please.
The next question is from Adrian Cighi with Credit Suisse. Please go ahead.
Hi there. Adrian Cighi from Credit Suisse. Thank you for the presentation and thank you for taking my question. I have a question on M&A and a follow-up on capital. On the M&A front, we've seen continued noise around potential M&A sort of opportunities in Italy. We've seen other banks provide a different stance on their intention, noting that shareholder value creation is their only criteria. Should they judge M&A as value accretive, they'll pursue it. You've previously ruled out M&A outright. Would there be any conditions where that would make sense? On the capital, sorry, just a clarification from your previous sort of answer. Can I just make sure that I fully understand the headwinds next year? You mentioned 20 basis points headwind on a net basis, including softer intangible this year and next year, 140 basis points between TRIM and credit risk migration.
Is that sort of outside of that headwind? This is sort of above and beyond what's already happened. Is that correct? Thank you.
Thank you very much. I will let Stefano confirm the regulatory headwinds. On the M&A side, let me reiterate the fact that, you know, our Team 23 plan is based on no M&A assumptions, only organic growth. It is even more true, frankly, with the current environment, where our clients are shifting to more remote banking. While in Team 23, we had a transformation plan, which is one of our four pillars. What we are doing today is to accelerate the transformation to more remote banking, to do what we were planning to do in four years, in two years. For instance, today in Italy, thanks to the transformation we did in the acceleration and to have a paperless branch and ability of the client to sign digitally, most of our products can be signed by our client remotely. Which allows us to handle much more remote banking.
We will debrief everybody at the Capital Market Day next year on the acceleration of the transformation. We prefer to transform rather than integrate, and this is driven large part by the behavior of our clients. The second reason why we have a no M&A stance is that we want to use our excess capital, as mentioned before, in order on one side to finance the economy. We see a rebound of the loan and the financing activity next year with a strong rebound of GDP to be expected. On the other side, to return capital to our shareholders. No M&A assumptions for Team 23. Stefano, I let you confirm the figures of regulatory headwinds for this year and next year.
Thanks, Jean-Pierre. Your summary is correct, so m inus 20 basis points from a common equity on ratio basis points impact from regulatory headwinds in the full- year 2020, and less than 1.4, so 140 basis points in 2021. Important to mention that in definition of regulatory headwinds, we're including both the effect deriving from models, change in regulation, and PD procyclicality as well. We are including all the related effects.
Thank you very much. Clearly, 70 basis point impact of procyclicality coming from the rating migrations are included in these figures, and were not included in our initial projection when we presented the plan, as we didn't plan for the COVID pandemic, basically. Once again, the rating migration impact, the 70 basis point, will afterwards be compensated when the GDP rebounds, the rating will improve, and we will get back most of it over time.
We also said, Adrian, that the 70 for the two years, 0.4 of that we've already taken year- to- date. Next question, please.
Next question is from Giovanni Razzoli with Equita. Please go ahead.
Good morning to everybody. A clarification on the regulatory headwinds, specifically on the PD migration, the 70 basis points. Can you share with us what geographies are mostly hit by this migration? Question number one. Question number two regards the cost of risk guidance for 2021, i f I'm not mistaken, the guidance for Italy should be in the region of 100 basis points for next year, if I'm not mistaken, if you can clarify this. I was wondering whether you have not recording any improvement in the risk profile because of the increased contribution from state-guaranteed loans. I was wondering whether you've kept a margin of safety vis-à-vis the improvement in the risk profile because of the contribution of state-guaranteed loans. Thank you.
Thank you very much. I will let Wouter give the breakdown of the projection for the cost of risk next year. TJ give the breakdown afterwards by country and the PD migration. Wouter on the evolution of our cost of risk this year breakdown between specific overlay and next year.
Okay, thank you. First of all, important to repeat that we confirmed the forecast that we have given for this year and next year. I clarify that for this year, we confirm the cost of risk, the stated cost of risk guidance of 100 basis points to 120 basis points. For 2021, we confirm the underlying cost of risk close to 60 basis points. The composition of our cost of risk, our stated cost of risk for this year, it's composed of three components. That's overlays, it's specifics, and regulatory headwinds. The specific cost of risk for the year 2020 is 40 basis points to 50 basis points, overlays 50 basis points to 60 basis points, and 10 basis points for regulatory headwinds. It's important to clarify that those 50 basis points to 60 basis points overlays, that is an anticipation of future impacts.
We have explained that in our methodology before, but it is important to highlight that it is very different from the specifics. When we go to next year, what we do is, we confirmed our stated cost of risk between 70 basis points and 90 basis points. When we correct that for the regulatory headwinds. That is the underlying cost of risk, which we confirm close to 60 basis points. That is a very important number because that is also what we use to calculate our underlying net profit for 2021, that has been confirmed in the range of EUR 3 billion to EUR 3.5 billion, and that is also the number that we use for our dividend policy.
Thank you very much, Wouter. Could you breakdown of cost of risk for Italy in 2021 projection and the PD migration by geography?
For 2021, we confirm our cost of risk overall at the bottom end of the range of 70 basis points to 90 basis points and close to 60 basis points. This is for entire group. We have not given the breakdown by division. In terms of the so-called PD migration for the group, already mentioned by Jean Pierre, 45 basis points is taken in 2020, and the remainder of the 25 basis points in 2021. Most of it is in Italy and as well as in Germany. In Germany, primarily because we have a lot of the CIB businesses, and we expect a rating migration drift from there, with some in Austria and very little in CE at this point in time.
Thank you. Next question.
Sorry, Jean Pierre. There was one question in relation to the state guaranteed.
Yes, go ahead, Stefano.
In relation to this, as a matter of fact, already in this quarter, if we look the dynamic of the risk-weighted asset, a positive contribution of around EUR 4.2 billion deriving from the state-guaranteed exposure. We have a positive impact on the expected loss perspective as well. You can see at page 41 of the presentation that our expected loss on the new business is moving down of a couple of basis points. This is also due to the benefit related to the state-guarantee contribution, that is two basis points when we look the contribution to the stock expected loss. If you look to the new business expected loss, when we look the contribution for Italy, it is even higher because it is 13 positive effect, 13 basis points to the expected loss of the new business of commercial banking in Italy.
Thank you. Also, Giovanni, the number Stefano mentioned on the state guarantees on RWAs on page 51. Next question, please.
The next question is from Andrea Filtri with Mediobanca. Please go ahead.
Yes, good morning, tw o questions. One on asset quality and one on capital. On asset quality, we continue to hear banks in Q3 experiencing very benign asset quality trends and no worrying signs from expiring moratoria, while economies are getting back into some sort of lockdown and GDP expectations are deteriorating. Can you share with us your latest thoughts and anecdotal evidence on what's happening on the ground on expiring moratoria, and what are really the forward-looking signs that we should be careful about or that you're seeing to gauge what real NPL creation is going to be next year? How will ECB look at this in conjunction with capital return for 2021? Thank you. The second one, capital return, a s has been said before, your MDA buffer is 2x your target at the moment.
If the ECB maintains the handbrake on capital return, how could you accelerate the usage of excess capital to boost your share price in a sluggish macroeconomic environment where loan growth is limited, also given your no M&A policy? Have you considered the possibility of buying back some of your JVs or other businesses within the group? Thank you.
Thank you very much, Andrea. First, on asset quality, if we go to slide 13 of the presentation, you can see that for the nine months to date, our specific cost of risk is at 33 basis points, which is below the specific cost of risk that we had last year. Clearly, in the current figures, there is no sign of a worsening of the credit environment, and for a large part, because of the moratoria and the fact that the different countries and governments have extended very massive government guarantee loans. If you go to slide 44 of the presentation, we gave the breakdown of our moratoria exposure and of the expired volume of moratoria. We have today EUR 29 billion of loans under moratoria, EUR 22.4 billion in Italy with EUR 1.6 billion, which have just expired.
The loans in Italy have just expired, so it is way too early to say if there could be any impact. At this stage on the loan expired in Italy, we have zero NP reclassification. I would say that is a very positive impact, but it is way too early to draw any conclusion. We had EUR 3.8 billion of loans expired under moratoria in the CEE side, where now we have a total amount of EUR 5.8 billion. They have expired a little bit earlier than Italy, and there we have seen a default rate, which is bigger than what we were expecting of, let's say, only 2%, 2% to 3%, depending on the country. It is, as well, too early to draw conclusion about where will be the default rate of the loans under moratoria lending, basically.
I think let's look at the situation step by step and not draw early conclusions about what can happen, but nevertheless, the early indications are more positive than negative, basically. You asked the question about what would the ECB do and how would the ECB look at the evolution of capital, you know. I'm at the ECB, and I cannot comment on their behalf. The only thing I can say is that, when the ECB looks, for instance, at share buyback, they look at the projection of capital under adverse scenario. On our side, we are always conservative. This is why we want to take additional provision in the fourth quarter to reach the 100 basis point to 120 basis point cost of ratio.
If you do the math and the calculation, it is an additional EUR 2 billion of provision in the fourth quarter to reach this amount, which is an anticipation, as we said, of regulatory headwinds as well as future defaults, so that this overlay provision, you know, it can be used against specific provision next year. I'm convinced that the regulator will do exactly the same. They will look at what should be the impact of banks, and if banks have properly provisioned or not their future regulatory headwinds or their future specific cost of risk. For that, we are very well prepared as we have a conservative approach, and you can absolutely, you know, project what should be our capital evolution. On the MDA buffer, it is high. That's very clear. It has never been so high at 438 basis points.
As we said, to project the excess capital above our 200 basis points to 250 basis points, you have to take into account risk-weighted asset inflation. We plan to have our loans increasing by EUR 10 billion next year, which will support our NII and allow us to deliver EUR 9.5 billion NII. I know we have to take into account the regulatory headwind. Next year, 140 basis points, basically, as well, and y ou have to project that. You know, we know the ECB will allow banks to pay dividends. It's not a question of if, it's a question of when, so i t will happen. We are convinced it will happen next year for the dividends on the 2020 net income, also for excess capital. For excess capital, we think that the ECB will only allow for a gradual return of excess capital.
This is why we plan only a gradual return on excess capital. We are not looking to do anything else. As we said, our business plan is based on organic assumptions, and that the no M&A amenity includes not doing anything on some of the transactions.
Thank you.
Next question, please.
The next question is from Domenico Santoro with HSBC. Please go ahead.
Hello. Hi, good morning. Thanks for the presentation, j ust a couple of clarifications. First of all, my understanding, if my calculation is correct, is that your guidance on net profit for this year, the EUR 800 million, which is EUR 100 million for Q4, implies a level of provision that is probably in the mid of your range, EUR 100 million , EUR 120 million or even less, around EUR 100 million , if we have to take some seasonality on cost, which is usual in the fourth quarter. I just wonder whether something can be better, actually, in terms of revenues, or if you might consider booking any capital gain in the case of a level of provision of EUR 120 million .
The second question is on the NII in Italy. Nobody asked, but it's down more than 10% if we have to include also the PCL3 impact. Apart from the business evolution, on which you commented very well, I just wonder whether there is any one-off here, because the decline is significant. A clarification on capital. Can you just mention what is left to book in the fourth quarter in terms of regulatory headwinds or positive, in the case of the software intangible, anything which is not business related? Thank you very much.
Thank you very much. Just on your first question for the EUR 800 million guidance for the full year, that's the guidance looking at our 100 basis point to 120 basis point cost of risk for the year, and we should land around the middle of the range, basically. That's the safest assumptions you could have, but that does include regulatory headwinds, basically, of 10 basis points. As such, the 100 basis points to 1 20 basis points includes 10 basis points of regulatory headwinds, which are not included into the underlying cost of risk. If we land on the middle of the range for the stated cost of risk, the underlying one should be at the bottom of the range of 100 basis points. It's exactly the same way that you have to use when you look at the 2021 cost of risk.
You said the stated is at the lower end of the 70 basis points to 90 basis points , which includes 10 basis points of headwinds. This is why we confirm the underlying cost of risk of close to 60 basis points for 2021. Once again, it's the underlying net income we confirm for 2021 of EUR 3 billion to EUR 3.5 billion, which is the base for our dividend calculation. For NII in Italy, we have some regular impact and one-off impact that we let Stefano give you the detail of that. Afterwards, TJ can comment on the last question on the regulatory headwinds capital. Stefano, for the NII in Italy.
In Italy, commercial banking in Italy, we have mainly three elements to be taken into consideration. We experienced an overall customer rate reduction on the lending stock of 12 basis points quarter-on-quarter, down to 226 basis points. It is the average client rate on all the stock. This was due to effect. One was connected with Euribor, that was affecting all the Euribor-linked asset. Second is a substitution effect, because we have a higher share of term guaranteed loans. The term guaranteed loans on average has a client rate above EUR 100 million, which were substituting the higher yielding short-term loans during this quarter. The second effect was a lower internal remuneration to the deposit due to the movement of the Euribor that was negative in the quarter for 17 basis point on the average.
The sum of these two effects is explaining the delta of the net interest income of commercial banking in Italy during the quarter.
More or less they account for half of the impact, basically.
Yes, the first one is equal to EUR 38 million, so the sum of the client rate effect, so deriving from both Euribor and the mix, and the second one is explaining the other delta.
Okay, TJ, on the question on the regulatory headwinds.
On the regulatory headwind, as we've mentioned sort of earlier, for 2020 is 20 basis points. For next year is 140 basis points, no worse than that. That includes PD scenarios, trends, and all of the effects in terms of models and proportionality. This is already everything that is factored in, that was mentioned sort of earlier.
Adding on this, you are asking for some more detail in relation to the fourth Q. The most relevant item, as I was highlighting before, is the software treatment or the changing of the software treatment is going to be in the fourth quarter. We are estimating the impact in being positive impact in relation to the common equity ratio, while the other one that is a negative is going to be some other effect from PD perspective. The sum of all the effect, as I said before, will bring the overall impact full year to less than 20 basis points.
Thank you. Next question.
The next question is from Britta Schmidt with Autonomous Research. Please go ahead.
Hi there. Good morning. Two questions, please. The first one, just picking your brain again on the wording regarding M&A. The statement that Team 23 does not include M&A is not necessarily the same as not being interested in M&A. Under what conditions would you be interested? Are there deals to be struck where you could maintain your payout policy, for example? The second one will be just on the stage 2 loans. It sounded like there were some single names in there. Can you just explain a little bit more what drove the Q- on-Q increase, and maybe also the single name, what sort of sector it was in, and what division it was in? Thank you.
I will let TJ answer for the second question. On the first one, you know, I just want to repeat what we said, and we always said, is Team 23 is based on no M&A assumptions. We prefer to transform rather than integrate, and we use our excess capital to finance the economy and to return capital to shareholders. Difficult for me to be clearer than that, basically, and don't try to find any subtlety in any of the wording. We never comment on rumors and speculation. It's not today that I will start to do that. No M&A assumptions, we prefer to transform than integrate, and we use excess capital to finance the economy and to return capital to shareholders. On the second question, TJ?
Thank you, Jean-Pierre. On the stage 2 increase, this is primarily driven by our proactive classification ahead of the moratorium expiration in 2021. We have basically gone through the portfolio, assessed the riskiness of the specific industry on a forward-looking basis, as well as the early portfolio warning. In Italy, we have roughly about EUR 5 billion increase. Overall stage 2 for the group is about 14%, Italy is still 9.3%. We are already looking at forward-looking, anticipating, just to avoid the cliff effect of the moratorium.
You were mentioning a single ticket. During the presentation, we were highlighting a single ticket contribution in relation to the coverage of the NPE. From this perspective, the effect was on the UTP. We experienced a shift of a position on a state guarantee position in Germany that we moved from performing to non-performing, that is explaining the majority of the delta of the NP related to UTP during this quarter. This is what I was referring to during the presentation. This is why we need to adjust the coverage of the third quarter to take into consideration the fact that the contribution of this classification is diluting the coverage of the UTP.
That would be a feature going forward for all banks. It's absolutely not UniCredit specific, which means that when you have a state guarantee loan, which is reclassified under stage 3, basically, we were going to provision only for the part which is non-guaranteed. Because of the way the coverage ratio is calculated, you take the full nominal of the loan, so it mechanically lowers the coverage ratio. This is why we said that if we were not looking or excluding this German credit, which was reclassifying UTP, our coverage for UTP would actually be above 50%, so higher than the coverage of the second quarter. We will give the two figures going forward for analysts to understand what is the true evolution of the coverage ratio, and m aybe just a small clarification as well, or detail on the stage 2.
On the stage 2, we have a group coverage of 4.1%. We have for Italy, a coverage of the stage 2 loan of 6.8%. Stage 2 loan in Italy represents 9.3%, as TJ mentioned, of our portfolio in Italy and are covered at 6.8%. The reclassification of 5 billion of loan in Italy from stage 1 to stage 2 includes additional NPEs of 100 million. Thank you. Next question, please.
The next question is from Jean-François Neuez with Goldman Sachs. Please go ahead.
Hi, good morning, and thanks for the call. I had a question on business development. You mentioned yourself that you prefer to transform, and you have a high priority to remote banking, paperless, et cetera. I just wanted to know whether you could or you had anecdotes or evidence how that has so far or not differentiate yourselves from your competitors in the various geographies where you operate, either in terms of volume, speed or pricing advantage.
My second question was again, maybe on some of the regulatory developments, which was, sub-question one, whether there is any correlation between the cost of risk that you anticipate and the procyclicality impact on your risk-weighted assets that you anticipate, meaning that if your early read on the anecdotal evidence on asset quality development, which you said yourself had been so far behind, even though it's very early, would have also an impact on procyclicality if they ended up having a positive impact on normal rates. Secondly, whether your TLAC excess, but whether in your NII projection, what assumptions were there on your excess of TLAC, and whether you plan to maintain such a buffer over and above your TLAC requirement, which is quite costly, obviously, as Group does. Thank you very much.
Thank you very much, Jean-François. Just on the remote banking side or transformation, we'll give much more detail at the Capital Market Day to be held next year. The reason why we want to have it in the first half of the year, not early in the year, is to show concrete acceleration of the remote banking measures that we have been taking. The teams are working very hard on that. I'll just give you one example. We have many, but we told the market that we wanted to be paperless in Italy from the beginning of the third quarter, which we are actually. But to do that, we have put in place what we call a digital mailbox, which allows the client to receive all the information on their UniCredit mailbox.
Using the digital mailbox, we can have the client today signing remotely for most of the contract, they can transact with us. Basically, moving from paperless in the branch using digital mailbox allows us to accelerate the remote banking activity by putting a lot of the contract into remote signing, for instance. That's very important, specifically when we look at the second wave, and we roll out the paperless process and digital mailbox in all our geographies. More detail at the Capital Market Day. On your second question, we encourage you to have gotten everything but the procyclicality impact that we are seeing. It's something which will impact all the banks, because the rating migration in the current environment, when you have GDP going down that much is mechanical, basically. It's not a UniCredit specific.
You know that we are always anticipating regulatory impact to put them behind us. This is why we are targeting an overall impact of procyclicality of around 70 basis points. We are taking 45 basis points of that this year and the balance next year, basically. As I said, this procyclicality will reverse starting from 2022, 2023 and beyond, and will be almost fully offset when the economy rebounds, basically.
Jean-François, do you want to clarify? Was it about the correlation between the cost of risk impact for reg headwinds and the capital impact for reg headwinds o r have we answered your question with what we just said?
That's right. For example, if you said 70 basis points, for example, next year of provision at the bottom end of the range. If you hit, for example, 10 basis points less of cost of risk for argument's sake, is there a mechanical effect on the reg headwinds with regards to procyclicality? I was trying to see what was embedded, if you want, or if there was any.
I will let Stefano give you the details, but the procyclicality is a risk-weighted asset inflation is not a provisional LLP. Stefano will give you more detail there, and afterwards we'll comment on the TLAC issuance of what we pre-funded this year and how we manage the buffer. Stefano, all yours.
In this moment, we have effect from PD procyclicality impacting the risk-weighted asset and clearly also the expected loss. This is deriving from the application of the internal models where we have two components, a qualitative one, and then a quantitative one. Looking to the qualitative one, we need to take into consideration the evolution not only of the macroeconomic situation, but also of the single obligor. This is not translating in the same approach that we are doing from the cost of risk perspective. As a matter of fact, we are not highlighting to you relevant effect on the cost of risk deriving from regulatory headwinds, because as we have communicated, the effect deriving from regulatory headwinds in the LLP is going to be around 10 basis points for the full year, mainly in fourth quarter and another 10 basis points in the next year.
As a matter of fact, there is no direct correlation between the calculation of the risk-weighted asset effect due to PD procyclicality and the cost of risk. For sure, the expected loss is going to be impacted accordingly. As a matter of fact, the vast majority of the cost of risk that we are having this year is due to overlay that are done looking to the specific sector and the specific position, especially one under moratoria.
If you go to 51 of the presentation, you have the risk-weighted asset world, basically. You have on the regulation, a EUR 3.3 billion increase of risk-weighted asset. This is for EUR 3.8 billion coming from procyclicality. We have also some benefit from SME guarantee. The procyclicality impact the risk-weighted asset, and as I said, will be reversed over time when the economy rebounds and the rating of the client improves. Of course, some of these clients might default in the future. That's part of the probability of default, which is a component of the expected loss. Procyclicality, the 70 basis points are purely risk-weighted asset-driven. Stefano, on TLAC buffer.
On TLAC, currently, as we can see at page 18, we do have an important buffer on TLAC that is 648 basis points due to different reasons. Primarily is the Common Equity Tier 1 ratio level that we have already commented is very high. Second, the delta between the fully loaded and transitional, that is equal to 74 basis points. A portion of this is going to fading away over time in the next year. Third, we've anticipated the 2021 TLAC funding plan for a couple of billion already during 2020. What we are expecting in the next years is an increase of the risk-weighted asset, also deriving from the regulatory headwind, that effect that we were commenting before. Two, the increase also of the requirement, because nowadays the requirement is 19.5, and we are going to move to 21.5.
As a consequence of that, we are expecting to have a reduction over time of the buffer. The buffer will be higher than our target one, that is the range between 50 and 100 during 2021, will move more to the upper part of the range, 50, 100 in 2022 and 2023.
It's a combination, if you want, for compression of the buffer, of the fact that we have regulatory headwinds to a certain extent, plus the transitional phase which is fading in. We lens towards the upper end of the target, and we have been pre-funding, as we always look for windows, in order to pre-fund when we have the windows.
Thank you. Next question, please.
Okay, thanks.
The next question is from Delphine Lee with J.P. Morgan. Please go ahead.
Hi. Good morning. Thanks for the presentation, so j ust two questions as follow-up. First of all, actually, I wanted to ask on fees. Your guidance, basically, for full year, implies an increase in the fourth quarter. I heard your comment on the financing fees seasonality, but, if you just would, do you mind give us a little bit more color on the other components, and also for next year, because your EUR 6.4 billion implies an increase which is actually larger than the GDP assumption that you have of 5%. Just wondering, if there are any drivers that we should be looking for. The second question is, just sorry to come back on capital and distribution.
Your comments are very clear on the 50% total distribution, but I'm just wondering, are you concerned that the excess capital potential buybacks that you could do next year could be at risk because of what's going on on the lockdowns and the macro in Q4? Thank you.
Thank you, Delphine, j ust on the fees for Q4, we had an extremely good October, basically. We are very comfortable that we'll deliver the fees for basically being able to reach the guidance, which is on EUR 6 billion, slightly below EUR 6 billion. A level of fees for the fourth quarter, which will not be very different from what we had on the third quarter, basically. A very good October, as you know, the new lockdown that we could have are more, let's say, pragmatic, we don't expect the dramatic impact we had during the second quarter. If Stefano has a bit more detail.
In relation to 2021, we are guiding EUR 6.4 billion. I think it's important to have a look at full- year 2019 as well, because 2020 is a really peculiar year. If you look 2019, our absolute level of fees was EUR 6.3 billion. We are guiding EUR 6.4 billion. It is an increase of just below one percentage point. Within the split of the fees, we are expecting to have, in comparison to 2019, a related higher growth on investment and transactional fees rather than on financing fees, also due to the fact that the financing fees are connected to the growth rate of the lending that we think is going to follow the rebound on the economy, and as a matter of fact, is going to be pronounced starting from the second quarter next year.
On the capital side, your question on is the distribution of excess capital at risk, as I said, I'm convinced that the ECB policy of allowing banks to return capital to shareholders, a mix of dividends or share buyback, is not an if, but a when. We are comfortable that the ECB will allow banks to pay back capital, basically, to shareholders, and the timing will have to be fine-tuned. We're very confident that it will start in 2021. Thank you.
The next question is from Patrick Lee with Santander. Please go ahead.
Hi. Good morning, everyone. Thanks for taking my question. I just have a couple of follow-up on your cost of risk guidance. I think for the full-year guidance, as you mentioned, you are expecting a sharp increase in the fourth quarter. If I just look at the specific component of it, I think on a year-to-date basis, you are running at around 33 basis points. If I, arithmetically, putting it together with your 40 basis points to 50 basis points specific charge for the full year, we are talking about a specific risk of maybe 60 basis points to 80 basis points in the fourth quarter. In the context of second lockdown, but also extension of moratorium, is there any other specific risk you see already that will see such a sharp increase from one quarter to another?
On the other hand, if the specific is lower, out of prudence, is it right that you'll still top it up to around 100 basis point for the group basis? I think relating to that, looking into 2021, your current overlay, let's say, is around EUR 1.7 billion, probably more than EUR 2 billion by the end of the year. What is the actual mechanics of turning this overlay provision into specific release? I mean, do you take it on a specific loan-by-loan basis, or do you take a more portfolio approach or strategically tapping into it? Within this 60 basis points to 70 basis point, are you penciling in any further overlays in that guidance? Thanks.
Thank you very much. I will let Stefano comment on the accounting methodology to be used on the overlay. On your first question, as Wouter mentioned earlier, we target for the full- year 2020, the level of specific cost of risk from the 100 basis points to 120 basis points or 40 basis points to 50 basis points, and for the overlay of 50 basis points to 60 basis points , while we have a 10 basis point of regulatory headwind. For the specific cost of risk, we have a proactive reclassification on a certain number of loans on which we will work in order to anticipate potential degradation of the credit. The increase we have on the specific will be coming from a proactive reclassification into UTP.
Mostly also we did, for instance, this quarter on the CEE side, we have the proactive reclassification under UTP of a certain number of credit, as we want to be conservative and anticipate the effect on the future. We always take conservative assumptions, and we always want to anticipate the pain. That's what we will be doing in the fourth quarter in order to have a " cleaner", if you want, 2021. On the reclassification, Stefano, on the mechanics to move from overlay to specific, basically.
First of all, in the overlay, as highlighted, we are including all the facts connected with capturing the forward-looking dynamic of our portfolios. We are including the change in the macroeconomic assumption that we already did in the first quarter. It is 20 basis points out of the 48 basis points overlay that were in the nine months. Then we have movement to Stage 2 portfolios deriving from analysis in relation to riskier sectors than others, taking consideration the impact deriving from COVID-19, and single obligor analysis, especially the one under moratoria. As a consequence of these two analyses, we are moving positions from Stage 1 to Stage 2.
To give you the flavor of the magnitude of the provision, if you look at the slide at page 48 where there is the breakdown by stages, you can also see that the overall provision that we have, in relation to performing portfolio, Stage 1 to Stage 2, is EUR 3.8 billion. If you look one year ago, it was EUR 2.6 billion. It's EUR 1.2 billion more in a year. How it's going to work? It's going to work that a portion of the Stage 2 portfolio is going to migrate to Stage 3, not all the portfolio, but a portion of the portfolio. When the portfolio is going to migrate, clearly a portion of this is already provisioned, another portion has to be brought to the level edge of coverage, either UTP or bad loans. We are going to have also the other way around.
A portion of the portfolio can also migrate to Stage 1. The combination among the two is going to let us arriving to the guidance of 2021 cost of risk, that it is close to 60 basis points underlying cost of risk.
Thank you, Stefano, and j ust to, you know, maybe summarize, the overlay provisions are not a black hole, so basically we are extremely careful to make sure that we can write them back if you want to reassign the freed-up amount into specific provision, and t hat requires a very fine accounting approach, as Stefano has outlined. If you want more detail, feel free to call the IR team or have a direct discussion with Stefano. That's for us an anticipation, and then you can deliver it. Next question.
The next question is from Alberto Cordara with Bank of America. Please go ahead.
Hi, good afternoon. My question is, first of all, on the cost of risk. We talked about the fact that these new state guarantee laws have a negative impact on NII, but clearly they will have a positive impact on loan losses. My question is, what would be your cost of risk next year without the state guarantee laws, particularly in Italy? Connected to that, what rate of default on a multi-year basis should we assume on these laws? I'm going back to a question that was very much discussed in the quarter, this 140 basis points regulatory headwind in 2021. This is not new news because it is the same guidance that you gave in 2022. We've been talking a lot about it, but you basically confirmed previous guidance.
The question to you would be, provided that you have a very high MDA buffer, you may not necessarily need to take any action, but is there anything that you can do to mitigate this negative headwind? Connected to that, I remember that you also mentioned a 60 basis points negative headwind for 2022, and if I remember correctly, around 30 basis points for 2023. Is this still in place in terms of your guidance or not?
The very final question is, can you remind us what is the impact on NII from your replicating portfolio in the quarter, and how should expected this to evolve? Thank you.
Sure, you know, I think that the IR team need to come back to you for some of the details you're requesting, because we can't deal with it in too much detail with everybody. I will try to address a comment on a more qualitative basis. I don't know if TJ can give more detail, otherwise, we'll let you call the IR team later. On the state-guaranteed loans, you're absolutely correct. It's NII negative to a large extent, as these loans are at a tight spread. As outlined by Stefano during the presentation, we have seen in Italy, for instance, that the clients taking the term loans state-guaranteed, have been paying back some of their short-term borrowings.
There has been a substitution effect, which is NII negative in the sense that the short-term borrowings were at a spread, which was almost twice higher than the spread of the long-term loans, specifically. It could improve the loan loss provision, I would say that, and that's the qualitative comment I wanted to make. If you had listened to Finance Minister Gualtieri earlier this week, I think over the weekend, he mentioned that it was likely that the moratorium in Italy will be extended from the end of January to the end of June this year, as well as the state guarantee. We might have more state guarantee loan, and clearly there could be more substitution effect between the bank loan and the state guarantee loan specifically for the sectors which are the most impacted by COVID-19.
We gave you the breakdown of the sectors of the segmentation for high impact to low impact. If we have more time, basically, some of these clients in the high impact sector might go for more state guarantee loan, which will de facto reduce potential Loan Loss Provisions for the bank. That should be a positive. Difficult to quantify, but we'll hand over to TJ, if he can give you a bit more detail on that.
On the state guarantees, clearly the impact LLP will get back, but our expectation for this year is quite not a lot, a few basis points. Clearly we have assumed in the default rate assumption this is artificially suppressed. As I said before, in Q2, that it was roughly 3.5%, but as shown in our observation is 2.2%, so it's artificially suppressed. We expect that all of this future impact for a cliff effect is really been articulated earlier, and Stefano's mentioned in terms of the overlay, the macro assumption, and also the proactive classification.
In relation to the replicating portfolio, in the quarter, we had a net benefit contribution to the NII from the replicating portfolio. We have execution on deposit or EUR 350 million. This is EUR 10 million better than the previous quarter. Reason being the dynamic of the rates, especially the floating rates, because it is a net benefit. It's the difference between the fixed rate and the floating rate. In future, especially in 2021, we're expecting a lower contribution because we are expecting that the floating rates are going to remain the same while doing the rolling. When we do the rolling, we're going to achieve lower fixed rate. The margin of the effect in 2021 could be ranging between EUR 80 million and EUR 100 million in comparison to the assumption of Team 23.
Such a delta is going to be reduced over time in 2022 and 2023 following the dynamic of the volume invested and the rates, and we are not expecting a significant difference in compared with the assumption of the contribution in 2023 versus Team 23 plan.
Alberto, we have given the exact details on page 55. That's the footnote one to page nine. There you have the replicating portfolio Stefano just mentioned. On the other ones, so the remaining open points on the guarantees, we'll come back to you, and that's a promise, not a threat. We'll update at the Capital Markets Day next year, the regulatory headwinds, but as you said, in planning in the communication we gave in December last year, more than 60 basis points of headwind in 2022, 139 basis points in 2023. We will have, afterwards, we have to take into account the mitigating impact of the rating migration, which will be a positive in 2022 and 2023. We'll give you the detail later, but which will mitigate basically the planned regulatory headwind as the rating of the portfolio will improve.
The credit quality will play positively starting from 2022. We'll give you that later next year during the Capital Market Day.
Okay. I appreciate.
Thank you.
Thank you very much. Thank you.
Next question, please.
The next question is from Jackie Ineke with Morgan Stanley. Please go ahead.
Hello there. I have a question from the credit side. First, thank you for your comments about treating stakeholders fairly earlier on, and bondholders certainly appreciated you paying the coupon on the CASHES this year. I've got a question on these bonds. The EBA recently came out with guidance on aspects of legacy instruments. At the moment, nearly EUR 3 billion of those CASHES have a value of 80% of core equity. The EBA guidance might suggest that after 2021, the other 20% might have limited regulatory value. I think even if we take the worst case and assign zero value to that 20%, I think the other 80% appears to be still very cost-effective core equity compared to traditional equity. I was just wondering if the EBA opinion has prompted a review of the CASHES or if you're now thinking about them differently in any way. Thank you.
No, we don't think about them differently. Stefano can comment about it, just to say that the CASHES, when the grandfathering ends, will be considered as Tier 2 for EUR 600 million, and that's it, basically. There's no question mark, I repeat, no question mark about the regulatory treatment of the CASHES at Tier 2 when the grandfathering ends.
I think the only point for clarification, you mentioned EUR 3 billion. That was the original issued volume. We're talking now only about the EUR 600 million that has not been converted with the restructuring in 2011 into equity. There's currently EUR 600 million grandfathered AT1 that, as Jean Pierre was saying, will turn into Tier 2.
Right.
Thank you. Next question, please. I'm sorry.
Sorry. In terms of what the EBA gave as an opinion, it would suggest that that cannot count as Tier 2 after the end of 2021.
No, I said there's no question mark. I cannot comment publicly about what the regulator says. If I'm telling you there's no question mark, there's no question mark, and the CASHES will be converted into Tier 2.
Great. Thank you.
We are absolutely certain of the regulatory treatment.
Perfect.
Thank you. Next question, please.
The next question is from Ignacio Cerezo with UBS. Please go ahead.
Yeah. Hello, good morning, Thank you for the presentation. A couple of ones for me. One on NII, if you can give us the calendar of maturities of your treasury portfolio in the next couple of years, and possibly the client of those loans expiring. The second one on the capital return, sorry to come back to that. There's been some chatter around the possibility of a payout cap imposed by the regulators. That's something actually you can reconcile with your conversations with them. Thank you.
Well, no, we never comment on rumors and speculation about UniCredit. I will certainly not comment about rumors and speculation about the regulator. The only thing I can say on the last question is that we are highly confident that the ECB will allow banks to return capital to shareholders. It's not an if, it's a when, and we're highly confident that it will happen sometimes in 2021. Also, on the NII.
On the portfolio.
maturities in the portfolio, let Stefano answer.
Yeah. The overall liquid portfolio is at EUR 136 billion at the end of the quarter. The duration of the portfolio is around 3.5 years, if you look at the overall portfolio. If you look the Italian component of the portfolio, the exposure is around EUR 43 billion, has a duration of 3.3 years. In relation to the maturities of these exposures, so the Italian one, we are expecting to have maturity around EUR 9 billion in the next year, and another EUR 14 billion in the year two and year three, coherent with the average duration I was highlighting to you, that is 3.3 years of the overall Italian government portfolio.
Just this Italian portfolio of EUR 43 billion, we have EUR 22 billion in the L2 connect, basically in the balance in the fair value of OCI.
Thank you very much.
Thank you. Next question, please.
As a reminder, if you wish to register for a question, please press star and one on your telephone.
Okay. It looks like there is no more questions. We'd like to thank you very much for your question. We will meet digitally with many of you in, you know, the next few days. Let me summarize some of the key points that we've been mentioning during this presentation. First, UniCredit is in a very strong position to face the future with confidence. Our management philosophy is very simple. We run the bank with a conservative approach, maintaining a tight control of cost and a strict underwriting discipline. We give priority to long-term sustainable outcome over short-term solution. It does mean in practice that in terms of net interest income, we will not compromise our future asset quality for short-term net interest income fixes. We, therefore, will not do volume lending, nor will we do carry trade. In terms of cost. Controlling cost is in our DNA.
As Stefano pointed out, Team 23 is based on substantial and additional cost-cutting over and above what we successfully delivered in Transform 2019, including further reduction of both FTE and branches. Less than a year ago into the plan, we have taken decisive actions on cost, and have thus been able to improve our cost guidance by a further 2% in financial year 2020 and 2021. As a result, we have improved our Team 23 gross cost-cutting reduction target by 25% to EUR 1.25 billion. In risk, we have a conservative provisioning and a strict underwriting discipline, which are the group trademarks. Our cost guidance is based on realistic and not optimistic assumption, and we always anticipate future impact. As a consequence, we are not changing our target because of the second wave.
Instead, we confirm our financial year 2020 stated cost of risk guidance at 100 basis points to 120 basis points, thanks to this anticipation of future impact as well as seasonality next quarter. This approach put us in a very strong position to face the uncertainty created by the second wave. We have an extremely strong capital position, underpinned by a record 538 basis points CET1 MDA buffer, and a very strong liquidity position with a point-in-time liquidity coverage ratio of 183%. We have de-risked our balance sheet, having materially wound down our gross NPE, sold non-strategic assets, and reduced our BTP exposure. As the successful execution of Transform 2019 showed, we have confirmed track record of delivering on our commitment to investors, whatever the environment. We confirm our 2021 underlying net income of EUR 3 billion to 3.5 billion, and our 2023 underlying return on tangible equity of 8%.
From this position of strength, we will continue to support the economy and distribute capital to our shareholders. We confirm the reinstatement of our capital distribution policy, based on an ordinary distribution of 50% of underlying net profit and a gradual distribution of excess capital when the regulator will give its green light. It will comprise a mixture of cash and share buybacks, and we plan to make both ordinary and extraordinary distribution from calendar year 2021 onwards. This concludes our third quarter results presentation. I look forward to talking digitally with you all again in a three months' time, if not before. Stay safe, and thank you very much. Bye-bye then.
Ladies and gentlemen, thank you for joining. The conference is now over. You may disconnect your telephones.