Good morning. This is the Chorus Call conference operator. Welcome, and thank you for joining the UniCredit Group First Quarter 2020 Financial Results Conference Call. As a reminder, all participants are in listen-only mode. After the presentation, there will be an opportunity to ask questions. Should anyone need assistance during the conference call, they may signal an operator by pressing star and zero on their telephone. At this time, I would like to turn the conference over to Mr. Jean Pierre Mustier, Chief Executive Officer of the UniCredit Group. Please go ahead, sir.
Thank you very much. Good morning to all of you, and welcome to our Analyst Call for the First Quarter 2020 Results. Before we start, let me make a few introductory remarks. This quarter was not a normal one for us or anyone else. The COVID-19 pandemic that is currently sweeping the globe has turned the daily lives of most of us upside down. It has resulted in healthcare systems of nation-states being brought to their limits, sometimes even beyond them. It has resulted in human tragedy and economic disasters for individuals or small businesses. It will have a profound impact on the global economy and may change forever the way we live, work, or interact. This is true as well for UniCredit and our stakeholders. As a result, this presentation will look different from our usual quarterly presentation.
For the first quarter 2020, we have added several pages dedicated to COVID-19 and the profound effect it has had on our group and on all its stakeholders. As we start a new financial reporting cycle, we have also simplified and streamlined the core part of the presentation. The overall deck contains the same information as before, but some of it has been moved to annexes or the divisional database. Moving on to the result, we had an excellent start to the year, with commercial revenues up significantly. The performance in the first two months allowed us to close the quarter with commercial revenues up 0.1% year-on-year, despite the COVID-19 induced slowdown visible from mid-March. As per our CMD19 guidance, there were some non-operating items in the quarter, such as the transaction reducing our Yapı stake, as well as the integration cost in Italy.
As a result, we have a stated net loss of EUR 2.7 billion, while the underlying net profit is close to breakeven at -EUR 58 million. Our capital is very strong. Our CET1 MDA buffer strongly increased to 436 basis points even after absorbing almost EUR 1 billion of impact from the update of our IFRS9 macro scenario. The core pillar of Team 23 plan will remain our strategic priorities. We will be updating the strategic plan and present our new assumptions when the environment stabilizes at a Capital Markets Day towards the end of this year or early next year. As I said in the beginning, this was an unusual quarter. We adapted the presentation to this situation. For information no longer in the core document, please go to the annex, where we have also added information on our ESG positioning. Let's turn to slide four.
These are key financial figures. Mirko will give you more details about this later. Our capital saw a sharp improvement with an end of quarter CET1 MDA buffer of 436 basis points, up 124 basis points quarter-on-quarter and 218 year-on-year. Mirko will give you the details later. As you would be expecting, on top of the additional LLPs from the IFRS9 macro, the P&L this quarter saw some impact from COVID-19 on revenues. They were down almost EUR 500 million in the quarter. The large part of that is explained by non-recurring items that are not linked to the operating profitability of the group. Examples include quarter-on-quarter changes in XVA for almost EUR 175 million negative, non-recurring valuation adjustment on participations like Visa for a total of EUR 65 million negative, or swing in the balance of other income, including Ocean Breeze, for a total of EUR 120 million negative.
Dividend went down quarter-on-quarter following our disposal activity such as Mediobanca. Our balance sheet asset quality remains strong, with our gross NPE ratio below 5% at 4.9% for the first time in many years. Let's turn to slide five. UniCredit was a pioneer in terms of responding rapidly to the challenges posed by COVID-19. We put, and will continue to put the safety of our employees and clients at the heart of everything we do. We were the first in Europe to close branches in a major way, starting in Italy. In branches remaining open to support clients at this time of need, we immediately provided protective equipment, including masks, hand sanitizer, and protective screens.
The majority of UniCredit team members rapidly transitioned to working from home with stable and secure access to all our systems. Thanks to this initiative in Italy, we were able to share best practices quickly with all the other countries where we operate, significantly improving our integrated response across every geography. Our IT teams are doing an outstanding job supporting group-wide working from home setup. In a few days, we were multiplying our remote capabilities by 15 times and secured 8,000 laptops before the worldwide rush to snap up such equipment. We have done our very best to support all our stakeholders in this unprecedented situation through various actions. We supported the frontline fighters in the war against the virus by sourcing and distributing protective equipment and respirators to hospitals, and also carrying group-wide fundraising activities.
We also partly sponsored the creation of the first prototype of the so-called CURA pod, which is an open-source project to create plugged-in intensive care units, ICUs, in converted shipping containers. I, and the top management, waived our financial year 2020 variable remuneration and dedicated the proceed to the fight against COVID-19. Thanks to the incredible commitment and hard work of all our team members, we have remained fully open for the business. We have continued to support our clients in these challenging times. We have completed numerous transactions for multinational corporates, raised funds in support of government agencies in their fight against COVID-19, and worked with client supply chain to help stabilize cash flow for SMEs. We have also helped our own suppliers in a similar way. Let's move to slide six.
Over the last three months, we have made more progress in our digital transformation than in the last three years. Our ability to do so is thanks to the significant investments made in our digital and mobile capabilities as part of Transform 2019. Our customers have rapidly and increasingly embraced digital solutions. Active mobile users are up 27% year-on-year, while digital sales as a percentage of total sales increased by 47% in the same period. This crisis has structurally changed client behaviors for both individuals and corporates. The adoption of multi-channel is accelerating and represents an opportunity to accelerate changes, leveraging on investments made in mobile banking, call centers, internet banking, and paperless branch. We will reallocate our investment priorities for IT and start framing to support such an evolution. Let's move to slide seven.
We were the first bank in Italy to offer moratoria to our clients affected by COVID-19 well before regulations were put in place. Following in our footsteps, many governments have now introduced moratoria by law. In total, we have provided EUR 28 billion of loans under moratoria. We work closely with all government agencies involved in COVID-19-related loans backed by state guarantees. In total, we have provided EUR 1 billion of loans backed by state guarantees so far. Guarantee processes have only recently been put in place by different administrations. UniCredit was the first bank in Italy to close a guarantee under such a scheme, and we expect the number to increase meaningfully in the coming weeks. Talking only for UniCredit, we expect the biggest volume to come from Italy. We expect to reach around EUR 15 billion of guarantees in Italy.
Such guarantees loan will provide much needed liquidity to clients and help save them this unprecedented situation, and consequently, will positively impact our cost of risk. Let's turn to slide eight. Our strategy is, and will remain so, to be a simple, successful, pan-European commercial bank. This lends great resilience to our business model. We have a strong presence in our 13 core markets in Italy, Germany, Austria, and CEE, where the expected impact of COVID-19 is markedly different. We are fully focused on delivering efficiency services to our 16 million clients, even more vital in this current operational environment. Our geographic and business diversification provides stability not only in the current environment. Thanks to our strategic investment, we are accelerating our digital transformation. Let's move to slide nine. We have and will always be focused on disciplined risk management.
We clearly demonstrated this during Transform 19, and it remains a core principle of how we do business. We took decisive actions in order to do the right thing to safeguard our shareholders' capital. To mention just a few examples, we reduced our BTP holdings by more than EUR 10 billion year-over-year. We prefunded most of our TLAC subordinated instruments early this year. We accelerated our non-core rundown and reduced gross NPEs by more than half while significantly increasing coverage. We sold non-strategic assets such as taking Yapı on our holdings in Fineco and Mediobanca, as well as real estate in Germany, for a total of more than EUR 7 billion. With 2020 hindsight, we chose a good time to do so. We shall keep that discipline going forward. There is no COVID-19 impact on our underlying cost of risk yet.
This is expected to evolve during the year. We'll give you more details later in the presentation. Now, let me hand over to our co-CFO, Mirko Bianchi. Mirko, all yours.
Thank you, Jean Pierre, and good morning to everyone. We had a strong start to the first quarter with excellent commercial performance in January and February. During the course of March, sales activity started to decline with the advent of COVID-induced lockdowns in all our countries. Our revenues are down EUR 472 million, or 9.7% quarter on quarter, of which, as Jean Pierre said, EUR 390 million are due to some non-recurring items not linked to operational profitability of the bank. Trading is down EUR 300 million, of which -EUR 174 million from XVA and -EUR 65 million from non-recurring valuation adjustments on participations like Visa. Dividends are down EUR 31 million due to disposals such as Mediobanca.
Balance of other income and expenses is down EUR 120 million, mainly from Ocean Breeze disposal. Our costs continue to trend down as in Q1 COVID related operational expenses remain marginal. Let's turn to slide 12. Below the line, in a non-systemic risk charges, it was a straightforward quarter. With regards to integration costs, we closed our Team 23 negotiations with the Italian trade unions as planned and consequently booked the integration costs after CMD 19 guidance. Let me remind you that we closed two transactions in Yapı shares during the quarter. Taken together, they took our indirect ownership of 41% to a direct ownership holding of 20%. This changed our regulatory consolidation from proportionate to at equity.
As a result, we no longer consolidate the pro rata risk-weighted asset of Yapı, but deduct the equity stake from our CET1 capital. The net impact of the transactions was + 58 basis points in core Tier 1 ratio. The P&L was affected mainly by the reversal of the negative FX reserve to the P&L, which was capital neutral. There were also some transaction-related expenses. Finally, we changed the segment reporting of the Yapı stake from CEE to Group Corporate Center, as it is now a non-strategic investment. Taxes in the quarter were negative, almost entirely driven by a taxable gain on real estate in Germany.
The normalized tax rate in the quarter was close to zero. Let's turn to slide 13. Let me remind you that at our last CMD in December, we introduced the concept of underlying net profit. This was done for strategic reasons. We wanted to ensure that the relevant and true profitability of the bank is clear and to show how it evolves. To calculate underlying net profit, we exclude non-operating items from stated net profit. This quarter, as per guidance, we had a total of negative EUR 2.6 billion of non-operating items that were excluded.
Details can be found in the annex at page 49. As you can see from the profit distribution across the group, we have a diversified business model. The performance of CEE stands out and underlines that this division is an important driver of the diversification and profitability of the group, as it contains many countries with a strong contribution, such as Romania. The post-tax impact of the IFRS 9 macro update additional group-level impairment was EUR 902 million. Without this, the divisional performance would have been different, and underlying group return on tangible equity would have been 6.5%. You can see the details in the footnote on the page. Let's turn to slide 14. Net interest income in the quarter was down 0.5%. The main driver of this were the customer loan rates, which contributed a negative EUR 48 million in the quarter.
Half of that is due to lower base rates in CEE, while the other half is driven by competitive pressure in Germany. There was also a tax-related positive one-off in Germany, which contributed +EUR 50 million to NII. Finally, let me remind you that the first quarter days effect was lower than usual because 2020 is a leap year. Let's turn to slide 15. Fees were up 5.2% year-on-year. The foundation for this strong performance was an excellent commercial start to the quarter in January and February across a number of areas. Commercial banking in Italy had one of the best first two months ever in terms of investment fees. CIB had one of their best quarters in debt capital markets. Post-mid-March COVID-19 has had a strong impact on a number of fee categories.
Gross AUM sales in the last two weeks of March, for example, were down low to mid-double-digit percentage points, depending on geographies. Levels in April have only recovered somewhat. Let's turn to slide 16. Overall trading income in the first Q2020 was EUR 165 million, down EUR 300 million quarter-on-quarter. We saw a solid performance in equities and commodities with an increase quarter-on-quarter of almost EUR 60 million, which was more than offset by XVA. The client-driven trading income, excluding XVA, was only down EUR 103 million quarter-on-quarter, EUR 65 million of which was due to non-recurring valuation adjustments, such our stake in Visa. The other trading income, which was down EUR 23 million quarter-on-quarter, there was a mid-double-digit impact from mark-to-market losses on government bonds from treasury positions. These are expected to recover fully.
The dividend line went down following the disposal of our non-strategic stakes in Mediobanca and Yapı Kredi. While this will result in lower dividend income in the future, the disposals strengthened our balance sheet and put us in a position of strength as to face COVID-19. Let's turn to slide 17. Costs continued to trend lower, both quarter-on-quarter and year-on-year. In the first Q2020, we regained some of the fourth Q2019 seasonality. As we said at our Capital Markets Day in December, in 2023, it's more about bottom-up process optimization. As a result, our cost efficiencies will be more back-end loaded to offset cost inflation, leading to overall flat costs over the plan period. These savings will fund the necessary IT spend over the plan period. COVID-19 had so far a limited impact on our cost base.
Year-on-year, we had EUR 5 million less of travel expenses, largely related to the current situation, while in the quarter, we had EUR 12 million in extraordinary COVID-19 related costs for branch refitting and remote working. Let's turn to slide 18. As you can see, our underlying cost of risk at 29 basis points is so far unaffected by COVID-19. We are even below our previous guidance of 34 basis points. Cost of risk is, as previously announced, estimated to be in the range of 100-120 basis points for the full fiscal year 2020. This will be a combination of the IFRS 9 macro update non-loss provisions we took this quarter and the expected recognition of sector and specific non-loss provisions throughout the year as risks materialize. The latter are likely to occur towards the end of the year, once the moratoria expire.
There were essentially no regulatory headwinds in the quarter. Let's turn to slide 19. With the introduction of Transform 2019, our loan origination is expected loss-driven. You can see the impact of that on the left-hand side, where the expected loss of new business is below the expected loss of stock for all periods. Also, both numbers steadily decreases over the quarters. Last year, we made an extra effort in light of the late business cycle and focus on new business on the best-rated clients. The result can be seen in the right-hand side. More than 70% of new origination and more than 60% of the stock have an expected loss below the average. Let's turn to slide 20. The shape of COVID-19 trajectory with regard to GDP remains unknown, with different expectations for each country.
In Western Europe, we're only just entering the start of the lockdown exit strategies, and it is too early to tell how things will evolve. As an illustration, we have shown here for a Western European country, two possible exit trajectories that differ by how long they remain at the low point of the lockdown. Our economists have estimated that an additional two months at the low point will cost six percentage points of GDP growth, which for Western European countries is massive. This is consistent with the latest commentary by the ECB, which expects Eurozone GDP to fall between 5% and 12% this year, depending on the trajectory of the low point of the lockdowns and the speed of the removal of the containing measures. ECB's 7% delta between the trajectories is very similar to our 6% delta.
As the sensitivity is high, it is of paramount importance that assumptions taken by the bank are realistic. Let's turn to slide 21. In order to estimate the impact of COVID-19 on our cost of risk, we have taken the GDP assumptions that you can see on the left. We have applied them to our EUR 485 billion credit portfolio, which you can see on the right. We have clustered the portfolio by GDP sensitivity and arrived at four segments from high to low impact. Only 10% of our loans fall into the high impact category, and these are the sectors most sensitive to COVID-19 headwinds, such as airlines, shipping, and tourism. For more than half of our portfolio, on the other hand, we only expect a low impact, including for mortgages of private individuals. Let's turn to slide 22.
As a first step for 3Q 2020, we have updated the IFRS 9 macro assumptions with our GDP growth rates. In ordinary times, we do this only in the 2nd and 4th quarter, but COVID-19 warranted an extraordinary update. We applied the updated macro assumptions to our performing portfolio, which resulted in higher probability of default and those increased loan loss provisions. These provisions for our performing portfolio amounted to EUR 902 million additional loan loss provision in the quarter on our loans. You can see that while the high impact portfolio comprises only 10% of our loans, it is responsible for 33% of the additional loan loss provisions. The low impact portfolio, on the other hand, results in a comparable amount of loan loss provision that is more than five times bigger. Let's turn to slide 23.
For the full fiscal year 2020 cost of risk forecast, we started from the first quarter that already includes the additional provisions from our IFRS 9 macro scenario update and added both sector-specific overlays and the individual provisions that we expect later in the year as risks materialize. This will occur as exposures migrate down the rating scale and may get classified as non-performing more often than not after moratoria expire. There will also be LGD effects. As you can see, total cost of risk is expected to be between 100 and 120 basis points, including the IFRS 9 macro, while the regulatory headwinds contributes less than 10 basis points as there is some time shift of models into fiscal year 2021. As seen previously, also for this analysis, the high and medium impact segments generate over 54% contribution to cost of risk, reflecting our conservative approach to the provisioning.
The provisioning for FY 2020 takes into account the migration effects of the announced government measures. Let's turn to slide 25. Our gross NPE stock for the non-core group was stable in the quarter. Our gross NPE ratio stood at 20.4% on our own definition and 2.8% using the EBA definition. This is almost exactly matching the average of the EBA sample of European banks, which is 2.7%. Let's turn to slide 26. The non-core rundown progressed well, even better than expected in Q1. That is traditionally a seasonally quiet quarter. Gross NPE in non-core were down EUR 0.5 billion to stand at EUR 8.1 billion. We are currently assessing the impact that COVID-19 will have on the NPE secondary market, and we'll update our non-core rundown strategy in due course. Let's turn to slide 27.
Our CET1 capital is at a very strong 436 basis points buffer over our MDA level. This is the result of two separate effects. On the one hand, we saw an organic increase of the absolute level of core Tier 1 ratio, thanks to the release of the fiscal year 2019 dividend and share buyback, as well as lower risk-weighted assets from the change in Prudential consolidation of Yapı. On the other hand, our MDA level decreased significantly, thanks to the application of CRD V Article 104a for 77 basis points as well as our lower SREP P2R for 25 basis points. We expect to remain well above our target range of 200-250 basis points CET1 MDA buffer throughout 2020.
The recent changes from the revision of the CRR and the ECB strong recommendations for the banking sector to use additional flexibility, such as, among others, moving to IFRS9 Phase In instead of fully loaded, will bring additional improvement to the CET1 ratio in fiscal year 2020. On a transitional basis, this amount to more than 0.8 percentage points, while on a fully loaded basis, more than 0.2 percentage points. Please note that there is a shift of around 0.5 percentage points of regulatory headwinds from fiscal year 2020 to fiscal year 2021, mainly following the flexibility rules recently published by the ECB in response of COVID-19, partially offset by the updated macro scenario. This is purely a time translation and does not change the overall amount. Let's turn to slide 28.
In line with the strong increase of our CET1 MDA buffer, our TLAC MDA buffer has increased to 391 basis points, well above our target range. This was driven by significant trade funding activity in the first quarter, pre-COVID-19, when we successfully raised EUR 4.5 billion of subordinated TLAC instruments at a very attractive level. Thanks to this, we have already completed close to 80% of the subordinated TLAC funding plan for fiscal year 2020. Let's turn to slide 29. Tangible equity stands at EUR 51.2 billion, more than EUR 2 billion higher than a year ago. We took decisive action in the first Q2020 to put the integration cost in Italy and the impact of the IFRS9 macro update behind us. As a result, quarter-on-quarter, tangible equity decreased by EUR 1.8 billion.
It was driven by the EUR 1.2 billion stated net loss, net of the AT1, and a decline in the revaluation reserves of EUR 1.3 billion from FX and securities that was only partially offset by a EUR 0.6 billion gain from the DBO. Jean Pierre, back to you.
Thank you, Mirko. Before I conclude and we go to Q&A, let me briefly talk about the easing of lockdowns and what we are doing. As governments across Europe start to lift the restrictions, we are ready for the so-called phase II. Just as demonstrated flexibility and speed when it came to the initial lockdown, we will apply the same principle in the coming weeks. However, we will base all our actions on scientific data, not dates. The safety of our people and clients remains at the heart of what we do. Remote working for central functions will remain in place for quite some time. Some of our people will be invited, not required, to come to the office, and we will listen and adapt to our people's needs.
We are confident that in phase II, we should be able to open 90% of all our branches in Italy and Austria, followed closely by Germany. Thanks to our strong multi-channel platform, we will continue to be fully operational regardless of how many branches are physically open. Let's turn to slide 32. Before we go to Q&A, let me reiterate the three key messages of this quarter. First, our business model is diversified and resilient. We have pan-European scale, 16 million clients that bank with us. We're accelerating our digital transformation, and we have a very strong capital base. These factors will help us ease the impact of COVID-19. Second, the core pillars of our Team 23 strategy remain. We will be updating the strategic plan and present our new assumptions when the environment stabilizes at the Capital Markets Day toward the end of this year or early next year.
Last but not least, we will continue to protect our employees, support our clients, and contribute to our communities. This is also the best thing we could be doing for you, our investors. Once more, our interests are completely aligned. Before taking your question, let me extend my deepest thanks and appreciation to all UniCredit team members whose commitment, resilience, and incredible hard work in this unprecedented situation has allowed UniCredit to prosper, enabled us to do the right thing for all our stakeholders. May all our employees, clients, and you, our investors, stay healthy and safe. I wish you the same for all of your loved ones. Now, Mirko, the rest of the team and I are ready to take your questions. If you could be please so kind and limit your questions to two each. Many thanks. Operator?
Excuse me, this is the Chorus Call Conference Operator. We will now begin the question and answer session. Anyone who wishes to ask a question may press star and one on their touch-tone telephone. To remove yourself from the question queue, please press star and two. Please pick up the receiver when asking questions to improve your audio quality. The first question is from Jean-Francois Neuez of Goldman Sachs. Please go ahead.
Hi, good morning, and thanks for the presentation. I would have therefore two questions, and the first one would be on asset quality, in particular, with a focus on your scenarios and the action on the moratorium. You're showing a balance of around EUR 28 billion of moratorium loans. A 10,000 ft view on that could argue that once they lift, they could all become potentially, or there is a risk that they don't return to, well, they migrate to non-performing. I just wanted to understand what's your strategy to make sure that this doesn't happen, and in particular, whether you think that there is a risk of non-financial, let's say, politically driven action as to make these loans maybe, or to extend this moratorium and make them less likely to stay performing.
Secondly, I just wanted to understand whether the cost of risk trajectory that you've given, on the other hand, what this would be in the base case of the ECB, GDP prediction. My second question is on capital. If Mirko has given some sort of update on regulatory headwinds and potential delays, et cetera. I just wanted to understand what's included in there, in particular with SME discount factor, infrastructure discount factor, software intangible, et cetera. The recently announced measures, and also whether the previously disclosed headwind, whether they change or not with regards to all of the measures that have been taken, in order to understand fully the work of the capital ratio going forward. Thanks a lot.
Thank you very much, Jean-François. Jörg?
Yes. Thank you very much for the question. Because we're doing this virtual for the first time, let me just see if I get this right. I think the first one on asset quality and the moratoria, was to say, how are we thinking of the risk content of that and what's our strategy to prevent these return NPE once the moratoria lift, as well as what's our view on any political actions for the moratoria to be extended. Then on the capital, the question was on regulatory headwinds, what's included of the recent CRR revision measures such as SME supporting factors, software deductions, et cetera. If our old regulatory headwind guidance that we gave at the Capital Markets Day is still valid. With that, I give it to Jean Pierre.
Thank you, Jörg. Let me make an overall comment and afterwards I hand over to TJ for most of the points. On the moratoria, we have taken specific actions on our side, and then we applied as well the government request that came. We have EUR 28 billion, as illustrated on page seven of the presentation. There is a specific case for the CEE of EUR 7 billion. As for EUR 5 billion, are in what we call opt-out countries. In other words, in specific countries, the government ask to have all the loans under moratoria, and clients who do not want to be under moratoria can choose to opt out. To clearly create EUR 5 billion of opt-out volume, and gradually the client can opt out if they want. EUR 2 billion are in opt-in, meaning the client choose to be in moratoria.
Besides CEE, the biggest impact is in Italy, where we have the EUR 19.4 billion, which comes from 100,000 clients. We have EUR 3 billion of mortgages and EUR 16 billion of loans to corporate. In this loan to corporate, we more or less have EUR 8 billion. 80% of the EUR 16 billion, which are for corporate of good credit standing, and 20%, so more or less EUR 3 billion, of corporate which are in the lower credit rating, so which might be a bit more sensitive, but TJ can elaborate a little bit more on that.
Before I let TJ comment more on the moratoria, I just want to point out as well, and TJ can comment about it, as you ask, what could be the cost of risk evolution in the best case of the ECB data we are giving sensitivity of the cost of risk in evolution based on the GDP evolution. You can see that in the annex, basically. TJ will comment about it, if the sensitivity at the current level of GDP, it is nonlinear. For 1% of GDP is between 4- 6 basis points, TJ will guide you through that later. I will let TJ comment as well on the capital regulatory headwinds, and the impact of the change of CRR. TJ, all yours.
Thank you, Jean Pierre. As Jean Pierre already mentioned, for the moratorium, we have seen a leveling off, firstly in terms of the number of requests. Already Jean Pierre has mentioned, of the EUR 19 billion in Italy, EUR 16 billion is to corporate client. 80% is good rating, so it is a good sign that a large part of the portfolio are in very good client. The EUR 3.2 billion of mortgages, the LGD is low, and these are really high-rated clients. We are closely monitoring the evolution of that, and where they are eligible for government guarantee financing, we will clearly help them to do so. On the capital side, clearly we will use ECB flexibility rule. There is a time shift of close to 50 basis points from 2020 to 2021. Clearly, we will monitor that. We have not factored in the so-called SME supporting factor.
From our discussion with ECB, they're very constructive to ensuring that they do not add more, either in terms of work burden or capital burden, at this Stage of the cycle. Thank you, Jean Pierre.
Okay. Next question, please.
Maybe just, I think on the regulatory headwinds capital impact, just to come back to the question of Jean-François, is the revision of the CRR and kind of anticipate a certain number of items. We put on page, there's some impact. Basically the SME supporting factor for us, which should be anticipated in the second quarter, should be around 19, 20 basis points, basically. That's something which was supposed to come later in 2021. It will come in 2020 on the second quarter. We have a certain number of further impact. The software exemption for CET1 deduction will be around 10 basis points as well. That's anticipated probably for the third quarter instead of 2020, instead of the end of 2021. These two factors on 30 basis points, which have an impact in terms of fully loaded capital, are moved forward to 2020.
There are a certain number of other items which have a low impact, and the IR team can comment about that if you need, on the bilateral basis. The ECB has been relatively vocal for asking the banks, and it's more a phased-in issue, to shift the IFRS9 into a phased in rather than fully loaded for those who were in fully loaded. As you know, we always want to take the most conservative side, so we're fully loaded. We will do that in the following quarters. That does have an impact in 2020 of 52 basis points, but that's a phased-in level, which will be reduced to 25 basis points in 2021.
This is why on a phased-in basis, between SME supporting factor, software, and IFRS9, we have an impact which is above 80 basis points, and on a fully loaded basis for this year, we should be above 20 basis points, but probably closer to 30 basis points.
Next question, please.
The next question is from Alberto Cordara of Bank of America. Please go ahead, sir.
Hi, good morning. My question is about taxes. I see that we have a profit before tax that is deeply negative, and yet you don't have a positive tax rebate. If you can explain that, if you can give us a bit of a guidance for what we should expect this year. Thank you.
On the tax side, I will let Stefano maybe, or Mirko comment on the tax impact and potentially guidance for the year. Mirko?
You're right. In terms of the tax base for the first Q, the tax base in Italy basically got affected by the one-offs, Yapı and integration costs. Therefore, the taxes were driven primarily by the gains in real estate in Germany. It's the sale of asset that we have done, and a normal, let's say, tax rate in the scene. We gave a guidance in Capital Markets Day of 18%-20%. At this point today, there's not enough visibility for updating our previous guidance, we're going to update as soon as we have better data in order to be able to guide you properly.
Thank you. Thank you very much.
Next question, please.
The next question is from Adrian Cighi of Credit Suisse. Please go ahead.
Hi there. Thank you very much. Thanks for the presentation. I have two follow-up questions, one on cost of risk and one on capital. On the cost of risk outlook, you expect over 200 basis points cost of risk for Italy this year, more than double the outlook of some of the other peers in the same geography. However, you have one of the highest coverage ratios and have, over the last year, de-risked the performing and non-performing loan book in Italy. Given the inherently opaque nature of the risk profile of your loan book or any loan book of a bank, what reassurances can you provide investors that this is a net consequence of some material difference in underwriting standards, but it's just sort of a more conservative take on the outlook? The second on capital.
You wrote back the capital to 37 basis points CET1 from full year 2019 dividends. Does management still expect to pay the dividend if the base case economic scenario plays out and if the ECB allows it? Thank you.
Thank you very much. Let me take that question, and I'll give you some hints on the cost of risk, and TJ will comment as well. On the dividend 2019, we will wait until the fourth quarter to see how the situation evolves and to see what is the ECB recommendation. At this Stage, we are in a wait-and-see position, and we will decide later in the year. On the cost of risk, you're right to point out that we have one of the highest coverage of actually any G-SIB in terms of the LLP that group level at 65%. In Italy, we have the highest negative GDP impact of anybody at this Stage at - 15%, which is an extremely strong impact, and of course, with an exponential impact when the GDP gets into more negative territory, specifically.
As you know, we want to be always extremely conservative in what we do. This is why we have, based on this -15%, the 200 to 240 basis point, but it's a natural evolution, if I may say, of the cost of risk because of the GDP evolution. With being said, to go back to your question on underwriting standard, and TJ can add more on that. We have shown in the presentation what we do in terms of the origination of the new business, you can see in the page that we have originated new business on the mostly investment-grade category in the coming years. It is translated into an expected loss, which is actually much better than the average expected loss of the portfolio.
If you go to page 19 of the presentation, you can see, for instance, that the expected loss on new business in the first quarter of 2020 was around 29 basis points, while the expected loss on the stock of the portfolio was 36. We are actually working very hard to be very disciplined, and 72% of the new origination in the first quarter of 2020 was with clients which are strongly into the investment-grade category. The team has instruction and is working to make sure that we shift the portfolio toward the good credit, and it's not a first quarter 2020 action. You can see that in the past, in the first quarter 2019, where the expected loss on the new business well below the expected loss on the stock as well.
We keep having a very disciplined risk approach, and that should translate into an improvement, as you can see, of the expected loss of the portfolio, everything has been equal. I don't know, TJ, if you want to add anything else.
Jean Pierre, hi. Thank you. I would add, clearly, we're taking a very prudent approach to the cost of risk assumption. You're right that the cost of risk expectation is around 200 to 220 for Italy, as you can see in the annex of slide, I think, 48. Just as a comparison, if we look back to 2008 and 2009 versus the current period, and here, like for like, clearly, we strip out the so-called non-core portfolio. There we have taken a look and see at that point in time, between the 2008 and 2009, the default rate went from 1.7- 2.5. Today, if you look at our assumption, the default rate in 2009 for Italy, ex non-core, was about 1.76%. Our expectation for this year in the prudent assumption is about 3.5%. This is almost double. Clearly, we're definitely taking a very conservative approach to the provisioning.
As Jean-Pierre's mentioned, that our asset quality has improved over the last few years on the very disciplined underwriting, as you have seen in the expected loss slide.
Thank you. That was very helpful.
As we said, we gave you on page 48 as well the sensitivity. Just let me be clear about the sensitivity for any additional 1% drop in GDP, the cost of risk group level should increase by between 4-6 basis points. That includes both IFRS micro-scenario as well as the specific provision evolution, and includes as well the impact of government guarantees, where the sensitivity is nonlinear, so that at around the current level. Operator, next question.
The next question sir is from Andrea Vercellone of Exane. Please go ahead.
Good morning. I've got three questions, but one is just a clarification, or it's a small detail. They are all on asset quality. The first question is on your guidance of cost of risk. It's a slightly qualitative question. You commented earlier on in the presentation that the big jump, or the jump, whether it's big or not, we'll see, in NPLs is back-end loaded to Q4 and next year. At the same time, you have the guidance that points to a cost of risk higher this year than next year, regardless of where the correct level will be. It's higher this year than next year. If you can just explain why that would be since the big jump in provisions is in Stage 3 loans, NPLs, and not Stage 1 or Stage 2. The second question relates also to page 48, but is a slightly different directional question.
You have the sensitivity saying for every additional 1% drop in GDP, cost of risk moves up by 4-6 basis points. You also point out that this is not linear. I'm more interested in the other way around. The delta is mainly in Italy, because that's where you have the big gap in lost GDP between 2020 and 2021. What if your assumption turns out to be too conservative, and GDP drops less, or the gap in terms of lost GDP is narrow? What does that do to cost of risk? The clarification is just on the new definition of default, which you were planning to implement in 2020, if I'm not mistaken. Is that still the plan for 2020, or does that shift back to 2021? Thank you.
Thank you very much. I will take the first question and let TJ answer for the second and third one. We have given a guidance for the cost of risk for the full year of 100- 120 basis points. We said that most of the default will materialize on the fourth quarter, because potentially we can expect that a lot of the loans which could default are today under moratorium. We should see the defaults at the exit of the moratorium more on the fourth quarter. This being said, behind the specific provisions to be taken in the fourth quarter, we expect to take a sector provision, which will then shift into specific provision when the default occur on the second and the third quarter, on the fourth quarter, which will anticipate default, which can happen later afterwards.
This is why we have a group cost of risk expectation of 100- 120 basis points for the full year. It's a combination beyond the IFRS9 generic provision already taken. It's a combination of specific provision, bulk of it should be mostly on the fourth quarter, and a sector provision which will anticipate the specific one, which can happen later. The cost of risk in 2021, which is between 70 and 90 basis points with the rebound of GDP we're seeing, and the lagging effect of the decrease in 2020. I'll let TJ comment on the second question on the sensitivity of the cost of risk and what happen if the GDP improve and on the new definition of default. TJ?
Thank you, Jean Pierre. Clearly, in terms of the cost of risk for the sensitivity, to the extent that our macro assumption is less severe than what we have built in, there will be a lower cost of risk impact. It shows that our assumption today is we are probably one of the most conservative, so we'll see. Too early to tell, but if the assumptions are lower, we will have, in essence, a write-back of some of the generic provisions, which we'll then monitor to see what kind of specific provision will come through. In terms of new definition of default, there's no change by EBA in terms of timeline. We are assuming today, for all of our assumption, that the new definition default will still go ahead as planned to be in place by January 2021. We're planning today for this to be implemented this year.
Just one point in terms of GDP evolution. The ECB mentioned that on the SSM, they will come up with guidelines in terms of GDP evolution for the year to be followed by various banks. Clearly, if these guidelines are more conservative than ours, we will adjust lower our GDP. If they are less conservative, we will keep our scenarios. We always want to be on the conservative end, we will keep our scenario and only adjust when the GDP fully changes in terms of figures. For us, it's important that we take the pain first, we take a conservative approach, and afterwards we should have only good surprise.
Thank you.
Next question.
Next question is from Domenico Santoro of HSBC. Please go ahead, sir.
Yes, good morning. It's Domenic from HSBC. Thank you for your presentation and hope all of you and your families are okay and well. A follow-up on the credit quality as well. My understanding is that you did some assumptions in terms of migration from Stage 1, 2 to 3, and that was basically the base for your cost of risk as well. Can you share with us what these assumptions are? You gave quite a visibility, actually, good visibility on the positive moving parts on the capital. Wonder whether we should expect any different direction in terms of risk-weighted assets at the end of the year when you're going to update your LGD. If there is any negative on this side. A question on the TLTRO.
Was wondering, what are your thoughts in terms of take-up in June, and what could be the impact at this point on the NII? Thank you very much.
Thank you. I will let Mirko comment on the TLTRO take-up, and then TJ comment on the assumptions and LGD impact we could have in terms of capital. Mirko, on the TLTRO first.
Yes. On TLTRO III, as you know, when we presented in 2023, the assumption was to basically repay the full allotment of TLTRO II. We have seen the much more favorable TLTRO III conditions that were announced by the ECB in March, and also recently on April 30th, that actually are easing banks' liquidity and cost of funding, sustaining new loan origination. What we're going to do is we are going to revisit that. In any case, the maximum we can take is about EUR 87 billion, therefore we are reassessing this.
By reassessing it, that if the conditions are good, we might be on the upper end of the maximum.
Exactly.
TJ, do you want to comment on the first two questions?
Yeah. Obviously, our so-called full year 2020 guideline in terms of cost of risk, is built into assumption on the flow to default, migration to the worst NPE status on top of the so-called Stage 3 LLP. Most of the IFRS 9 are mostly on what we call the Stage 2, and the rest is spread across flows to default migration on top of specific LLP.
On LGD migration and capital impact between TJ and Mirko. TJ, what do you see in terms of LGD migration, basically?
Well, at this Stage, it's too early to tell on the LGD sort of migration, because of the moratorium, there's little impact in observe. We've clearly built into our assumption on the flow to defaults and the migration to the worst debtors. That's reflected in the capital and the LLP assumptions.
We might have one in 2021, maybe some adjustment coming from the PD on the regulatory capital headwinds. Actually, something to be looked at a little bit later, basically. Next question.
Maybe we forgot to answer two parts of the question, sorry. One, you ask also what is the potential NII implication of TLTRO III. It's going to be north of EUR 300 million, if we take the full kick up. On the risk-weighted asset, the development, from a capital perspective, we'll be very comfortably positioned in our 200-250 basis points in MDA buffer. Of course, risk-weighted asset will slightly go up, but you know that the amount of guaranteed business will have a neutral capital impact, and therefore we need to now follow the mix, and how the mix will go through the various quarters.
Yeah, very comfortably, Mirko mean that we will be well above for the year of 200-250 basis point buffer.
Yes.
That should be the case as well for 2021. The capital side will remain very strong in this phase. Next question.
The next question is from Giovanni Razzoli of EQUITA. Please go ahead.
Good morning to everybody. Two quick questions. The first one, can you share with us what is the default rate at the Italian perimeter, that is including not only the commercial bank in Italy, but also the part of loans of the Corporate & Investment Banking that are based in Italy. The second question, very broadly, you've been quite detailed into discussing the impact on the cost of risk in 2020 and 2021 and also the sensitivity, which is actually quite low because 1 percentage point of GDP, you said, is the 4-6 basis points of higher cost of risk, which is a good number. Can I ask you what is instead, in general terms, the impact instead on the revenue? Once this overall situation is hopefully solved, can we assume that there will be a structural impact on the fees and NII?
Sooner or later, the cost of risk will normalize at 2021 or 2022 or so. My point is, shall we also factor in a structural contraction of the NII of the fees? Specifically on the fees, can you help us to understand what could be the impact on the overall activity in terms of volumes, traffic on the branches in the beginning of the second two? It's an unprecedented situation, and the factor in the lockdown is impossible. Any comment would be much appreciated, because my understanding is that also, we should not also forget the impact on the revenues other than the cost of risk, where there has been a lot of elaboration on your side. Thank you.
Yeah, let me take the second question, and we'll let TJ answer your first one. First of all, we benefit from a very diversified business model as you have seen, and we explained that and showed that in the presentation where we showed a breakdown between Italy, Germany, Austria. We have seen in different countries a very different client reaction, basically. That's because of the lockdown has been put in place differently. The resilience of the medical system has been different between the different countries. For instance, in Germany, in terms of AUM fees, we saw very, very little impact of the lockdown in the months of March. It was true in Austria as well. While we have seen an impact in Italy, which was much more meaningful.
As was mentioned before, January, February, we were at 150% in terms of investment fees compared to the average of 2019, and we went down in March to 50% of the average of 2019, and Germany and Austria remain more stable. With the phase II, we can expect Germany, Austria, and which on a combined basis represent in terms of loan, the same amount as Italy. We have the breakdown on page eight. We can expect Germany and Austria to rebound more than Italy, and CEE, which has EUR 66 billion of loans, so half of Italy, to rebound as well as CEE has been less impacted. That means that we could have a differentiated evolution based by the country.
This being said, for 2020, as we don't have a good understanding of how the lockdown will be lifted in the different countries, we don't want to give any guidance because it's more a bet than something which can be confirmed, and I don't like to say things on which I don't rely on hard evidence. It's different for the cost of risk. The cost of risk guideline we give are based on the almost mechanical application, let me say, of the GDP evolution that we gave, taking a conservative stance. Don't expect us to say something which can be too positive. We just want to remain as we are, always conservative, and say what we do and do what we say, period. In 2021, 2022, we should see an evolution of the GDP.
We have, as far as the European GDP is concerned, a + 10% rebound of GDP in 2021 versus a 13% decrease in 2020. Italy is - 15 +9 . Clearly the GDP will close the gap to trend slowly, and we expect this gap to be closed more quickly in Germany and Austria than in Italy, from the figures we gave, and probably the gap to be closed in 2022, 2023. We will see an evolution back to GDP trend in the coming few years. This is why we said during the wider call that if we look at the net income, we're not going into the detail of fees, NII, et cetera, because it's still a bit too early to give a precise indication.
If we look at the gap to trend of GDP versus the evolution we can have in our various countries, we think that we should be in 2021 in terms of net income, taking into account the LLP provisioning that we have indicated, 70-90 basis points. We should be at 75%-80% of the net income we were planning in Team 23. 75%-80% of the net income planning Team 23 was around EUR 4.3 billion, EUR 4.4 billion, gives us a net income between EUR 3 billion-EUR 3.5 billion, which, in terms of ROTE, should be around 6%-6.5% ROTE. That's the guidance we can give for 2021. We don't give any guidance of the revenues for 2020 because it's a situation which is, for the moment, not stabilized enough for us to be too precise. TJ, maybe on the default rate?
Yes, on the default rate for the so-called Italian parameter, which include both commercial bank in Italy, and the CIB component in Italy, is relatively stable compared to the Q4 at around 1.9%. Remember, in our projection, we have assumed that this is almost double to around 3.5% take into account the effect of the so-called government guarantee scheme. We have taken a fairly conservative approach in our default rate assumption for this year.
Thank you. Next question.
The next question. Excuse me, sir. The next question is from Britta Schmidt of Autonomous Research. Please go ahead.
Yeah, hi there. I've got two questions, please. One is just coming back to the assumptions around the macro. The cumulative GDP over 2020, 2021 of - 6% in Italy is not too dissimilar from what we see from other banks, still the 200-240 basis points is way above what other banks are guiding to. Do you think it's management conservatism baked into the translation into Stage 3 loans, or do you see in your internal modeling a potentially very wide range of outcomes as well? My second question would be, there has been some press reporting on the merchant acquiring business as to whether you could potentially be doing something with that. Could you give us a comment on that? Maybe also let us know how much it contributes to the P&L. Thank you.
I don't think we disclose the detail of merchant acquiring contribution to the P&L. What I can say that it is a business that we like. We said there is no M&A for UniCredit, whether in terms of combination, no disposal, our parameter is stable. We like merchant acquiring, and we will not do anything. With being said, we don't comment on rumors and speculation. That's to answer to your second question. On the first one, TJ can complement what I'm doing, but what is important is not only basically the overall net GDP impact over the two years, but it's also the trough. If you have a 15% negative GDP in 2020, the impact on the cost of risk will be much bigger than if you have a -7 or 8% .
This is why we take these conservative assumptions as being a bigger trough, and of course, a rebound, which net gives you a convergence which might not be very different, maybe from a lower trough and lower rebound. It is more conservative, and as such, impacts more the cost of risk, knowing that we have taken into account, of course, in our assumption, the positive impact as well of government guarantees. TJ, anything else to add?
Yeah. If I could add, if you look at page 20 of the slide, clearly, we're assuming that the lockdown is going on for longer two months. That in itself will impact the assumptions in terms of recovery. We never fully have recovered, and this will feed into the IFRS 9 sort of modeling assumption. That means that we will be taking a lot more provisioning in 2020 versus less in terms of anything generic. This will help us anticipate, if any, in terms of the so-called specific LLP that will come, particularly as we approach right after moratorium in Q4.
I think what is important as well, when you make the assumption, this is why we give you the breakdown on page 21, is to see how the portfolio is composed. This was already commented, the breakdown by segment, high impact to low impact, and clearly the high impact for us, which is 10%, has a very high contribution, not only for IFRS 9 provision in the first quarter, but for the full year. You have to look at what is the provision on one side, and what is the breakdown of the portfolio between the different segments to estimate the LLP.
Yeah. I would just add for the high impact, even though it's 10% of the portfolio, for the year, if you look to page 23 of the slide, we are allocating 39% in terms of so-called both IFRS 9 and factor sort of specific generic provision.
Thank you.
Thank you very much.
Next question, please.
Next question comes from Hugo Cruz at KBW. Please go ahead, sir.
Hi. Thank you. Two questions. First, can you please split your full year 2020 cost of risk guidance between Stage 2, migration to Stage 3, and top-up to existing Stage 3? Second, given the new environment, can you do better on cost-cutting compared to previous targets? Thank you.
Well, let me comment on the cost side. I don't know if we can give the volume Stage 2, Stage 3, but I will let TJ comment on that. On the cost side, we have demonstrated in Transform 2019 that we take always a very proactive and decisive actions on cost. We had a plan to keep our cost stable in 2023, an increase in 2020, 2021, then decrease to go back to a stable cost as a lot of the transformation has more back-end loaded impact. With what's happening now, a certain number of development have been put on hold, basically. We target for 2020 to have a cost base which should be stable versus 2019, basically.
If situation change, then we will change, and what we want to make sure is that we can, within our cost base, reallocate some of our budget towards what will help us transform the business model more quickly. As we said, in terms of IT priority, we will refocus a lot of our project towards full transformation, making sure that we broaden a fully digital offering for the product and improve even further the remote advisory side. I'll let TJ comment on the Stage 2, Stage 3 breakdown if we give it.
Yeah. Clearly for Stage 2, Stage 3, firstly, I think for the guideline for full year 2020, it's built on our assumption in both the macro IFRS 9 and factor that we look to on the close to default towards an NPE status, which means that some part will go from Stage 1 to Stage 2, and some will go from Stage 2 to Stage 3. This is built up into the 100- 120. Clearly, we have not given sort of the exact sort of breakdown, and we'll be happy to revert on this to our IR.
We can say that the IFRS 9 one is mostly on Stage 2, basically, and the rest is spread around -
Yeah
- for the default. Yeah.
Yeah. I think the factor specific will be sitting Stage 2 to Stage 3. Yes.
Yeah.
Hugo, we'll get back to you with additional information. Next question, please.
The next question is from Andrea Filtri of Mediobanca. Please go ahead.
Yes, good morning. Just following up on the question on costs. What are your considerations? I hear you are now targeting flat costs for 2020. How much of the therefore additional cost cuts do you reckon will be structural because of a different way of working for the group, and therefore, can we bring some of this forward, or most of this is just savings from people working remotely and not traveling, and so on? Secondly, on risk-weighted assets, what risk-weighted asset impact are you expecting from procyclical changes in PD and LGD, and how much of this will be cushioned by a negative volume evolution and the government guarantees? Finally, I just reiterate Adrian's question on the sensitivity of cost of risk to the upside of GDP, given it is non-linear. Thank you.
Sure. I will let TJ answer on the sensitivity of the GDP evolution as well as on the impact of possibly the PD LGD. On the cost side, as we said, we expect a flat cost in 2020. Clearly, the clients are changing, and the adoption of remote banking has increased de facto. If I may say, we expect that the clients will stay in a much more remote banking digital interaction with their banks going forward. That could help us simplify the process and simplify as well the business model. This being said, the cost savings that we could have in terms of process optimization, as we said in Team 23, are more back-ended because they are a consequence of process simplification, should appear more in 2022 and 2023.
We are accelerating some of these changes. That has already an impact in 2020, keeping the cost flat. We will apply Team 23, which was already anticipating this changing. It's not like it's something new because of COVID-19, it's just an acceleration for some of the impact. We'll get the benefit maybe earlier than what we were planning in Team 23. This therefore is no change in terms of what we expect in terms of FTE efficiency or branch closures versus what we were planning as we had transformation as one of our four pillar. On the risk-weighted asset and sensitivity, I'll let TJ comment.
In terms of clearly risk-weighted assets, the model recalibration, while we expect clearly the PD deterioration will impact the recalibration, we expect this to be offset by the benefit we'll get from the so-called loans that will move towards the state guarantee scheme. Overall, we don't expect any meaningful impact on the PD LGD negative migration. In terms of sensitivity, I already mentioned earlier, for 1% to the extent that our assumption is too conservative. Let's say by 1%, we expect the cost of risk to improve by 4- 6 basis points.
Next question, please.
The next question is from Antonio Reale of Morgan Stanley. Please go ahead.
Hi. Thank you for the presentation. I have two questions, please. The first one is on the government initiative. I guess, from your position as a pan-European bank, how effective do you see the government measures across some of the key countries you operate in? Any early issues or concerns you can share from your conversations with SME and corporate clients? I'm particularly interested in Italy and Germany, please. The second question is, you've been de-risking your earnings, and that was a key part of your business plan. You've done lots of work on the non-core and also have been reducing the BTP portfolio. How do you manage the large demand for credit and at the same time defend your marginality and asset quality? I've heard your comments on TLTRO III, but also how should we think about your BTP portfolio going forward?
I saw the reduction over EUR 10 billion year-on-year. There's talks about changes to the solid focus. I want to hear your thoughts. Thank you.
Sure. I will let the co-CEOs of Western Europe comment on the client activity on Germany, Italy, and government guarantees, and the process. I will take the second questions afterwards on the de-risking. First, if Olivier or Francesco want to comment about Italy or Germany as far as the SME side is concerned.
If we start from the Italian side, we believe in general that the schemes are significant and effective. Therefore, they should have the ability to make a difference in allowing our corporate clients to go through the most difficult part of the economic slowdown. We are very active on all three schemes. We've been rapid in receiving a large number of requests on the 25,000 100% guarantee amounts issued by Fondo, with 43,000 requests already processed to date. Essentially, a minimum refusal rate. That should be important in sustaining that small business, self-employed part of the clientele. The larger component will be Fondo Centrale di Garanzia run by MCC. We're currently 3,000 processing requests. We look to a significant increase. As mentioned by Jean Pierre, we've been first off the run.
Such a guarantee has issued already to, and we have several others in the pipeline, some of which may already be closed this week, expecting, as mentioned, over EUR 15 billion of overall amounts of guarantees in Italy.
Olivier, do you want to comment on Germany?
Sure, Jean-Pierre. What we see in Germany is, first of all, a great confidence in the values inaudible schemes, which were already in reality in place with the sponsored loan. We have an engine, if I may say so, that is working already, that has been working very well in Germany. As we can see, for example, for the small business, the initiative on the Schnell Credits for amounts below EUR 800,000 is working effectively very well. What we can see in terms of reaction is that it's a strong adoption with the value model and the great confidence on the ability to bridge the situation.
Maybe on your second question on the de-risking. First of all, on government bonds, we said that we will reduce our BTP portfolio to 50% of our tangible liquidity by 2023. We have reduced by EUR 10 billion, from EUR 53 billion, EUR 54 billion- EUR 44 billion within one year. We keep reducing, letting the portfolio naturally amortize. We have a duration, as we have said, which is around a 3.5 year as of the first quarter 2020 for the banking book, down from 3.6 years on the fourth quarter 2019. This EUR 44 billion portfolio is for EUR 23 billion in the to collect category and for EUR 21 billion on the fair value OCI, basically with sensitivity on capital on the after-tax basis, which is around 1.7 basis point from memories for 1.4 basis point for 10 basis point, a move of the spread.
On the credit side, basically we will have more demand for credit to go under the guarantee. That has, as mentioned earlier, positive impact in terms of risk-weighted asset on one side and cost of risk on the other. We are taking into account the benefits coming from the government guarantees on our cost of risk. Clearly, the loans under the government guarantees will be at a spread, which is going to be lower than the natural spread to a certain extent. That should be slightly NII negative, but the TLTRO on the other side will put in just a benefit, as illustrated by Mirko. A lower spread coming from the client spread, if I may say, because of the government guarantees, but improvement in the NII coming from the TLTRO, which hopefully should compensate each other.
In terms of the origination, we maintain clearly a strict risk discipline. At the same time, it's important for us to support the economy and when we can assist our client with a government guarantee, we do that. Government guarantees move from 100% guarantees. For instance, in Italy, on the EUR 25,000 loan, we had close to 30,000 requests. As of the 30th of April, we had already processed and validated 52% of that, so around 15,000 requests. We're trying to make sure we move quickly to support the economy, to support the client, and at the same time, maintain the right management of our risk profile.
Thank you.
Thank you. Next question, please.
The next question is from Azzurra Guelfi at Citi. Please go ahead, ma'am.
Hi, good morning. One question is on non-core. Will you confirm the target of non-core fully run off by 2021, given the changes in the market condition? The other one is on capital return in 2020 onwards. You have a strong buffer, and assuming that the regulatory and market condition normalize, and you are allowed to pay it, is the capital buffer a bigger driver than the profitability of the group that could be potentially lost in 2020? Thank you.
Thank you. On the non-core, I will let TJ comment on the current situation. I would say that there is no reason at this stage to change our forecast to run off fully the non-core by the end of 2021. TJ can give you the dynamics of the evolution. On the capital returning, clearly the environment is very different from where we were at the end of last year or early next year. We said as far as the dividend side is concerned for the dividend 2019, that we will wait for the fourth quarter to look at what is the economic situation and prospect of economic evolution, as well as the recommendation of the ECB. From 2021 onwards, we will look at the evolution of the situation. We maintain the guidance that we gave in terms of % payout, which was given in Team 23.
We will see if we have, depending on the ECB recommendation, how we manage excess capital, which, as I said we anticipate to be well above the 200 basis point-250 basis point buffer that we gave in Team 23 in 2020, but also in 2021. All that will depend on the prospect of economic evolution. TJ, do you want to comment on the non-core?
Sure. For the non-core, clearly, we are still aiming for full run-off of the book by end of 2021. We're progressing on all initiatives. Clearly, COVID-19 has an impact on the recovery, particularly those that involve judicial, because all the courts are closed. For the so-called unsecured portfolio, we're continuing to bring transaction to the marketplace. Three portfolio are now actually in the market. Actually, as of yesterday evening, we received for one of the portfolio binding bids. It is still open, clearly for the so-called unsecured portfolio. For the secured, where you require due diligence on the secured asset, this clearly will have to wait until when the restriction is lifted. Overall, we're progressing on all initiatives, and we still expect to run down non-core by 2021.
I just would like to add that the coverage ratio on the non-core with the additional provision we took on the fourth quarter is at a very high 78.4%. We are of the very strong opinion that we are provision to sell. The market, as TJ mentioned, still remains open. We maintain our guidance that the non-core will be run off by 2021. If you look at the net exposure of the non-core being provisioned at 78%, we basically, on the EUR 8 billion that we have left, the net exposure is less than EUR 2 billion. I wouldn't say EUR 2 billion are irrelevant, but you can see that it is now a very, very small amount. That's why we focus on the group X non-core, as far as the focus on the risk profile of the group is concerned.
Thank you. Next question, please.
The next question is from Delphine Lee of JP Morgan. Please go ahead, madam.
Hi. Thank you for the presentation. I just wanted to come back on asset quality. Looking at your slide 21, with the sector breakdown, would it be possible to have the sort of the asset quality metrics on the high impact portfolios in terms of NPE ratio and coverage? On asset quality, if I may ask, out of the EUR 28 billion of loans which are under moratorium, in your assumptions in terms of full year guidance, cost of risk 100, 120, sort of how much losses are you assuming against that EUR 28 billion of loans under moratorium? Then just one quick question on capital, just to clarify. It looks like, you don't expect much credit rating migration, and there are definitely some positives in terms of impacts on capital by year-end from regulation.
Just wondering, is there anything we're missing in terms of capital bridge by year-end, or should we expect significant increase by year-end? Thank you.
On the capital side, Mirko can comment in more detail, we said that we will have, in addition to the structure we have today, everything being equal, we'll have additional contribution on a phase-in of more than 80 basis points of CET1. That's the recommendation, I would say the strong recommendation of the ECB to IFRS 9 to phase-in for IFRS 9, which we will do well in a fully loaded. The change of the CRR, which will impact Q2 and Q3, as I mentioned earlier, on the SME supporting factor and on the software side, to combine more than 80 basis points CET1 positive impact for the year on the phase-in and more than 20 closer to 30 on a fully loaded.
We said as well that we expect to be well above the 200 and 250 basis points buffer in 2020, and I would say in 2021 as well. Basically, even if in 2021, as TJ mentioned, we will have the regulatory headwinds that we were planning in 2020. There has been a time translation as well, the one we were planning in 2021. The combined impact of regulatory headwinds for 2021 will be around 100, 120 basis points, basically from what we see. On the EUR 20 billion of loans under moratorium, as we said, we gave the breakdown between what we have in Italy and in CEE mostly. In CEE, as we said, there is opt out and opt-in country, so the opt-in is EUR 2 billion out of the EUR 7 billion. For the rest, it's more mandatory shift towards a moratorium.
In Italy, as we said, for the moratorium, I will let TJ comment about what we expect if we have the breakdown of the losses. We have on the EUR 19 billion, EUR 3 billion which are coming from a mortgage, so the LGD should be very small for any, obviously, if they default. On the corporate side, which is for the balance. We have 20% of the corporates which are in the lower credits where we could have a higher impact of provisioning, but the rest is in good credit quality, which should be fine. This being said, on the capital side, Mirko, any additional comments on capital evolution?
Yeah.
TJ can elaborate more on your asset quality. Mirko first.
As Jean-Pierre said, we expect strong capital position. The biggest driver is on one side to show the CRR relaxations, and most of it is coming from IFRS9 transitional. There is some risk-weighted asset dynamics that we developed over the rest of the year. Maybe that's the building block that you are missing in your forecast, and nothing else that is major aside from movements of federal debt for CR effects and so on.
TJ, on the moratorium -
Yeah.
- then on the NP coverage -
Yeah.
- metrics for the high impact. Yeah.
Yeah. Firstly, on the moratorium, already Jean Pierre mentioned the breakdown. Just one point is that 80% of, for instance, a corporate that has requested a moratorium are actually in a very good rating category. All of this clearly a part of our so-called bottom-up assumption in terms of the cost of risk forecast. For the sector, we clearly do a detailed sector impact, and on page 20, you've seen that 10% of the so-called portfolio, we've clustered them in high impact, with the transport, travel, airlines, shipping, tourism, oil, and auto suppliers. We clearly have that breakdown firstly by countries and within the country, what are some of the percentages. We do have the asset quality metrics, and we can follow up through IR just to give you all those metrics. It's too detailed to put it in.
Yeah. We'll be in touch.
Thank you very much.
Thank you very much, and next question, please.
The next question is from Christian Carrese of Intermonte. Please go ahead, sir.
Hi, good morning. I have two question, one on cost, one on capital remuneration. On cost, I was wondering if the COVID-19 crisis could delay little bit the phasing of the exit of the personnel you agreed with the trade unions. The second question on capital. Compared to last CMD, we are expecting now lower earnings but higher capital in a certain way. You said 80 basis points on a phasing basis already 2020. In terms of dividend payout, 40%, I suppose that it will be difficult to ask regulator to approve a buyback. In theory, you have more capital, you're going to pay lower dividends based on a lower net profit. I was wondering, compared to last December, if we look at the banking sector, we see that some banks have lost almost 50% of the value.
There are already some banks at 80% discount to tangible equity. This year, maybe, if you assume that the GDP will rebound by 10% next year, it's not the right time to maybe look at other banks, smaller banks to consolidate the sector. Maybe also the system would be in favor on this move, and use the bad will of those banks to clean those banks as you did with the non-core bank for UniCredit. If you can share your view on this. Thank you.
Sure. Let me first comment on the FTE exit. We have agreed with the Italian union, on one side, a very socially responsible action in terms of early retirement for our colleagues in Italy to around 5,000 net FTE reduction. This net FTE reduction include as well an agreement to hire 2,600 people. We have, as part of these 2,600 people, already confirmed that we will transform the temporary contract of some of the apprentice into a full-time contract. We have a plan for FTE exit and early retirement, this plan will be applied and could actually maybe be slightly accelerated. It is a voluntary request of our colleagues, I assume in the current environment, some of our colleagues might be willing to retire earlier than later, basically.
We shall see, but we keep our plan in terms of FTE management, and that's fine. In terms of the capital side, we said no M&A, and we said that, we repeated that, and we keep it. If I may say, even more today than before, it is extremely important to focus all the management attention towards the transformation. The last thing you do if you manage a bank today is to get more FTEs, more branch, and more integration when what we need to do is to transform much more quickly. We have a zero interest to enter into M&A transaction domestically in Italy or somewhere else, as we want to focus 100% of our attention in the transformation. Okay. Think that's clear, and we will not change that.
Thank you.
Next question, please.
The next question is from Ignacio Cerezo of UBS. Please go ahead, sir.
Hi, good morning. Thanks for the presentation. If I can follow up on the importance of the government loan guarantees, especially in Italy, if you're expecting a big portion of your corporate book to end up being rolled over into the guarantees, I'm meaning referring to the back book rather than new lending. The second one is on trading. How quickly do you think you can come back to your EUR 300 million, EUR 350 million run rates you had in the previous guidance? Thank you.
Thank you. On the trading side, we have a guidance which was closer to EUR 300 actually, more than EUR 300- EUR 350 at the last CMD. I think, with the current environment, the activity is client-driven for almost all of it. The trading activity is a consequence of the client activity, and we think that while the month of April was a good month for the CD side, on the revenue side, we will still have, in Q2, some impact of the XVA as it moves based on the market environment. If you take out XVA, which is non-recurring, the month of April in terms of client activity show a rebound of activity, and we think that we should go and be back towards a more normalized activity probably in Q4. Q2 and Q3 should be transitional quarters, and Q4 should be a more normalized one.
As far the government loan guarantees are concerned, we said that we expect to target more or less EUR 15 billion of government guarantees. That's mostly in Italy. If you look at our commercial banking Italy book, it's EUR 130 billion, and most of the guarantees will be for the commercial bank in Italy. You can see that it's a bit more than 10% of the overall loan book.
Thank you.
Next question, please.
Excuse me, sir. The next question is from Patrick Lee of Santander. Please go ahead.
Hi. Good morning, everyone. Thanks for taking my question. I just have one question on the moratorium, also one on your cost of risk guidance. Firstly, on the moratorium, am I right in saying that any loans under moratorium now is still considered performing and therefore accruing interest? If so, I guess that interest will be added at the back end as extra balance for the customers. If that is correct, can you give us some sort of a rough estimate in terms of the increase of the size of the interest burden for the corporate and for the retail customer calls the moratorium? Secondly, on the cost of risk guidance for the rest of the year, you have pretty much gotten to the same level as you have seen in first quarter at around 100 basis points.
I just want to check with you, is it fair to assume that under the IFRS macro assumptions methodology, and given your already very conservative stance, is it fair to assume that most of the macro assumption changes is already taken in the first quarter and consequently for the rest of the year, your guidance, is that assuming a pretty big jump in specific risks that you're expecting for the rest of the year? That's it for me.
I will let probably Stefano Porro comment on the accounting treatment for the loans under moratorium, but there is no P&L impact per se for the loans under moratorium, and Stefano can give you a bit more detail on that. On the cost of risk, I'm not going to understand completely your question, but if the IFRS 9 generic provision, they will change if we change our GDP assumptions basically. If they worsen, we'll take more provision. If they improve, there will be a provision release. That's mechanical. We shall see. You know that it is a three-year average, basically, of the GDP evolution over three year, which has to be taken into account.
I think what Patrick meant was actually because we take all the macro in the first quarter, is it fair to assume that for the rest of the year it's more a jump in specific LLPs?
Yeah, that's right. Yes.
Yeah. I think what I would also add, this assumption of clearly we'll be building up sector-specific provisioning as well, which is our assumption in terms of generic. Then we will see, post the moratorium, as the so-called specific provisioning comes through in terms of classification, then we will release the so-called generic to be into the specific.
Okay. Just to clarify, IFRS 9 on the macro basis is taken, and we don't expect to take more as we are conservative on the generic basis. On the provisions which are remaining, if you can see, we have under 100- 120 basis point, 20 basis point on a full year basis coming from the generic, the balance coming from the rest. It's a mixture of sector as well as specific provision. We want to anticipate some of the specific name provision for sector provisions, which will be reversed afterwards into a specific provision when a defaults happen, but to make sure that we can take this 100- 120 basis point fully in 2020 to put that behind us, basically. On the accounting treatment for the loans and the moratorium, Stefano, do you want to comment on the way it is handled?
Yes. Thanks, Jean Pierre. For the last part of the moratorium schemes, we have a payment suspension. Considering that we have no payment cancellation, the payment is always suspended, the bank will keep on accruing the NII in the P&L. From a customer perspective, it's a postponement of the cash flow, depending on the future of the private or public moratorium, the cash flow is going to be paid by the customer at the end of the waiver or at the end of the cash flow profile connected with the package, meaning either the private or the public moratorium.
Thank you. Next question, please.
The next question is from Anna Benassi of Kepler. Please go ahead, madam.
Good morning. My question is a clarification on the non-core division. I hear what you said. You confirm the run-off of the non-core, of the EUR 8 billion of the non-core, let's say EUR 2 billion net, as you said, by 2021. Is it the EUR 4.3 billion gross target for end 2020 confirmed? I heard that you said full portfolio up for sale, probably they are worth more like EUR 4 billion. Do you still see interest on this portfolio, and I think you commented already that the extra provision you took in Q4 you think is enough, also considering some price pressure in the NPL market. This is my first question. My second question relate to the contribution to systemic funds and the banking levies that are more like the EUR 900 million for UniCredit.
Do you hear anywhere some indication that some of them could be suspended for some time or removed given the situation, or we should stay with the assumption? Regarding the shareholder remuneration, given in Q1 you are running, let's say, break even, do you believe that the high capital buffer would be enough for the regulator to allow you to pay any dividends or launch any buybacks, even without a positive profitability or large net profits to cover your policy, and that applies to 2020 results, but even more to 2019 suspended distribution policy. Thank you.
Sure. Thank you. Just on your first question on the non-core, TJ has already commented, but we have said that we maintain the full run-off by 2021. I think the 2020 evolution will depend on the speed of the easing of the lockdown. Why? As TJ mentioned, while on the unsecured loan, we continue to work, and we received, as he mentioned, some bids for some of the portfolio. For the secured one, there needs to be some due diligence, and some of the due diligence need to be physical due diligence. As long as the lockdown is blocking people from being able to travel, that might delay slightly some of the transaction we want to do.
Just to clarify, on the non-core reduction, we were planning to have on the EUR 8 billion that we have left, more or less half coming from a disposal and the balance from recovery and write-off. Basically, we're not looking to have EUR 8 billion disposal, but only half of it. We are confident that by 2021 we run it off, but maybe based on the evolution of the lockdown, the EUR 4.3 billion might keep a little bit towards 2021. We confirm the target. On the systemic charges, we have no indication at this Stage that there will be a suspension or a relief of any systemic charges. We shall see if something happens, but for the moment, I don't think we should bet on anything.
On the shareholder remuneration, just to remind you that we were planning in 2020, on the 2019, to pay a dividend which is based on the underlying net income. Which is different from the stated net income, and because, or thanks to our high capital buffer, basically. Based on the situation, we will reassess what we do on the fourth quarter. On terms of economic evolution and indication of the ECB, it is way too early to be able to anticipate, one, the economic evolution, two, what the ECB will recommend banks to do. We will follow, of course, the ECB recommendations.
The final question is from Benjie Creelan-Sandford of Jefferies. Please go ahead.
Yes, good morning, everyone. Just a quick clarification from my side. I just wanted to be clear that the guidance earlier was for net profit in 2021 of EUR 3 billion-EUR 3.5 billion. If that is the case, just on the basis of the 70-90 basis point cost of risk guidance, the comment you made around costs and systemic charges, et cetera, we should therefore assume you're basically saying that the revenue run rate in 2021 should be close to a 2019 level, quite a sharp recovery versus the run rate in the first quarter of this year. Just checking on that. Thank you.
You heard me well when I said that we are targeting net income based on the current economic evolution. If the GDP evolution is different, we will have a different evolution, which should be between 75%-80% of what we were targeting in Team 23. Clearly, we will have an impact of higher provisions, as you pointed out. If we're at 70%-90% basis point of provision, it's an impact versus 40 basis point of provision that we were planning. The cost there should be in line or lower than what we are planning for Team 23. The revenues will recover from what could be expected for 2020, but will still be lower than what we were targeting for Team 23, that you round up to the guidance that we give.
There should be, in 2021, basically, net income would be based on the evolution of the spread. If the situation improves, there should be a strong recovery on the fee side. That's very clear. Normalization of the trading profit. I'm not going to go into more detail about it because it's way too early to go into the detailed mechanics of what should be the 2021 net income made of. Our expectation is more based on closing the gap to GDP and where we will be. The good news with a crisis is that they end at one Stage, and you go back to normality. Within our assumption, we start going back to normality in 2021 with a lot of unknown about the virus, the vaccine, the treatment.
I'm not a medical specialist, so for the moment, give you what is my best expectation, looking at the overall macro side, but which can change if the medical side or the virus evolution is different from what we could expect.
Thank you very much. I think, Jean Pierre, it was important, like Benjie was saying, that the numbers we gave are for 2021.
That's correct. Yeah.
2023.
That's correct. Okay. 2021, yeah. That's important. This looks, for now, to be the last question. Should we give a short time if there's additional ones?
As a reminder, star one for questions.
If there's no other question, thank you very much for participating to this call. We will have a discussion with some of you during the virtual roadshows that we're starting right after this call. I hope that you and your families will be well, and I look forward to meet you virtually or in person in the near future. Thank you very much, and have a good day. Bye-bye then.