Deutsche Pfandbriefbank AG (PBB) Earnings Call Transcript & Summary

Good afternoon, and a very warm welcome from Garching to the PBB Full Year Results Call. We thank you for joining us today. Here with me is Andreas Arndt, our CEO; and Andreas will guide you through the presentation and be available for your questions afterwards. So we'll stick to the proven concept. Andreas, please go ahead.

Yes. Thank you. Also from my side, welcome to today's analyst call. It's a sort of anniversary, if you put it this way, because it's exactly a year ago that we had our last year's annual results publication. After that, a few days later, saw the onset of the outbreak of the COVID-19 pandemic-related lockdown. And since then, 2020 has been dominated by COVID-19 pandemic with far-reaching consequences for people, for their social lives and for our economy and, of course, for the global real estate sector in general, and for PBB in particular. Now in the first quarter '21, we see some cautious signs for an upturn. Vaccinations are picking up. The easing of lockdown measures is being discussed. And an overall return to positive growth rates is expected. Nevertheless, we will stay cautious and prudent. Even though GDP growth, hopefully, turns positive in '21, we expect the real macroeconomic impact on real estate to become more visible only during the course of this year. Many companies are strongly hit. Further layoffs are to be expected, and structure changes like more home office or working from home, video conferencing instead of extensive traveling, positive experience with online shopping and all that will influence the demand for real estate space, the occupancy rates and, consequently, prices risks and yields in '21 and thereafter. And this is why we stay cautious. Our consistent risk conservative business approach paid out well in last year, and we want to keep it like this also for this year. Better be disappointed on the wrong side. Better one loan less and less margin than one loan too much with full LLP. Thus, it will be important to continue our careful selection of clients and properties and to keep disciplined focus on prime business while we have every intention to cautiously augment our core business despite difficult times and while we are continuing to build our investments into the digital future of our business. Digital future of our business with the launch of PBB's customer portal this quarter, where we deliver a fully functional customer portal for commercial real estate financing for our clients in Europe and the United States. Secondly, we started a comprehensive exercise in digitalization of our customer credit process, which will provide significant efficiency advantages for our customers as well as for the bank. And finally, with CAPVERIANT, we have already built meanwhile well-established open digital credit brokerage platform. ESG, in particular, the financing of sustainable green property -- of green properties is the next sizable challenge for our society. We will also address the structural change through respective investments. The first visible step was the issuance of green bond 4 weeks ago. 2020 should have given ample proof of what we are able to do, deliver risk-conservative transactions, generate solid results and, at the same time, continue to invest in our future. As such, you may call 2020 the test case in an extremely challenging market environment. Both the management of the credit portfolio and the existing business on one hand and the cautious selection of new business on the other were, likewise, important. We have intensified our already strict portfolio monitoring and further tightened risk parameters for new business. Thereby, our portfolio has so far proven its high quality. And in the new business, we reached good volume while increasing gross margins. Despite all uncertainties and in the light of COVID-19, PBB shows a solid operating performance with a PBT of EUR 154 million after EUR 216 million in the previous year, based on strong NII, stable operating costs while building up significant provisions for further potential impacts of COVID-19 in '21. That brings me to Page 4 of the presentation with a few comments ahead of the main core, so to speak. NII being 5% up year-on-year from EUR 458 million to EUR 479 million, mainly driven by lower refinancing costs as well as improved floor income. In addition, of course, NII also benefits from the participation of TLTRO III. General admin expenses have remained stable at EUR 204 million, almost stable after EUR 202 million last year, with higher personnel expenses especially driven by IT in-sourcing and ramp-up of capacities for regulatory projects, which has been mostly compensated by reduced non-personnel expenses. Investment in our strategic initiatives are financed intrinsically by continuous efficiency improvements. In the context of COVID-19 pandemic, risk provisioning increased, as I mentioned, to EUR 126 million from EUR 49 million last year, with most parts, i.e., EUR 70 million or 55% related to model-based additions in Stage 1 and 2, and that's reflecting recalibrations of all risk parameters based on worsened economic and sector-specific aspects and forecasts. Additions to Stage 3 worth EUR 57 million were driven by further revaluation of U.K. shopping centers, which were already provided for pre-corona. You know the cases, I think, very well. We have no new cases with need for Stage 3 provisioning, although 2 defaults, but as I said, no need for Stage 3 provisioning. I'll come back to that point. New business volume reached EUR 7.4 billion after strong Q4 with EUR 3 billion to account for, which reflects our selective approach in the light of COVID-19 and significantly lower investment and transaction levels in the commercial real estate markets. Average gross interest margin is significantly up by 25 bps to now 180 basis points. The strategic real estate portfolio remains stable at EUR 27 billion, supported by lower prepayments and higher prolongations. This said, our NPL volume remains on low level with EUR 470 million after EUR 510 million, as such, slightly down year-over-year with an NPL ratio of 0.8%. And for the sake of completeness, Public Investment Finance and Value Portfolio are noted at somewhat lower portfolio levels, according to plan, EUR 5.8 billion for Public Investment Finance and EUR 11.4 billion for the Value Portfolio. Funding remains stable. As already reported, pre-crisis activities provided us with comfortable liquidity position throughout the uncertainties in 2020 and even beyond, supported by TLTRO III, which allows for some funding optimization by substituting some other liquid covered funding, i.e. some more expensive repos. During the year and also early in '21, we had taken opportunities in the funding markets, both through constant and reliable private placements and, amongst others, the successful issuance of, in British pound, Mortgage Pfandbrief in benchmark format, which is, I have to say the first Continental European SONIA-linked issuance in the market with attractive levels, which we did execute in Q3. In total, we collected a total volume of EUR 3.6 billion in capital markets at stable average funding spreads versus last year. In addition, we issued EUR 1.4 billion own use Pfandbrief as collateral for TLTRO, which are not computed or not noted in the EUR 3.6 billion, just given the above. Meanwhile, spread levels have come down back to precrisis levels, which indeed provided for a strong start into 2021 with 2 successful benchmark issuance in January, i.e., USD 750 million Mortgage Pfandbrief and EUR 500 million inaugural Senior Preferred Green Bond. Capitalization remains solid at CET1 levels of 16.1%, reflecting stable RWA level and slightly increased CET1 capital due to the reduction and the expected loss shortfall given the uptick or the buildup of provisions. As much as we can, at this point in time, we want our shareholders to participate in this good performance and intend to make the dividend proposal to the AGM on the basis of the maximum allowed demand according to ECB's latest recommendations, i.e., $0.26 per share or 35% payout in 2020. If and once the restrictions by the ECB will be further eased or removed by end of third quarter, the bank will reassess its dividend options in the fourth quarter of '21. We see ourselves as dividend stock. The resumption of dividend payment in 2020 should be seen as an indicator that we stick to our overall dividend policy of 50% regular plus 25% supplementary dividend if and when and as far as possible. Given the ongoing and persisting uncertainties in 2021, we refrain from providing a defined range for profit guidance. However, we think to be able to deliver stable operating result in 2021. Therefore, we aim at a PBT in excess of last year's figures despite further risk provisioning and continuing or continuation of our investment program. In short, we remain cautious, but we also remain forward-looking -- positively forward-looking. Let me turn to Page 5 with this short overview on a couple of figures. First of all, I mentioned already new business, which were lower than in the year before due to selective approach and significant overall reduction in global transaction volumes. However, due to careful management of prolongations and visibly lower repayments, the bank managed to keep strategic real estate financing volumes stable with increasing its portfolio margin. The other point is operating income from lending business, NII remains robust while we keep operating costs under control. I think it's noteworthy that the cost-to-income ratio also in this year remains at a very competitive level of 42%. We did significantly build up risk provisions over the last 5 years, providing a solid platform for future potential impact from COVID-19. We'll spend some more time on this topic later on. All in all, and even after significant risk provisioning, PBT remains on a solid level. And if you sort of add back the risk provisioning, we achieved the best result from operating lending business since IPO. This brings me to some more details on the corona situation in '20 and '21. What you find on Page 6 is an overview of more operational aspects of how the bank did cope with the situation. I'm not going through all of that, and I will leave it for your reading, but I want to pick out a couple of aspects, which I think are important to take along. The first thing is, from day 1, the bank has been fully operational across all areas, at all times. We can rely on a state-of-the-art IT landscape, which allowed for remote working from the first day. Up to 90% of that has been and is working from home without any loss in diligence, accuracy and motivation. This includes also the critical functions of the bank. The senior COVID-19 crisis team has been implemented already in March '20. In this context, a weekly monitoring system has been implemented for IT and staff-related topics with a set of numerous risk indicators and limitations. Furthermore, we have implemented far-reaching prevention measures at an early stage. I just [ quote ] office presence, restrictions, team separation, social distancing and all that. The core of our efforts were, of course, all the measures around risk management. Portfolio margin was identified and our risk parameters for new business tightened. A Corona-Task-Force was established in March comprising several working groups along the -- and across the divisional lines and covering specifically affected clients or assets as well as more technical topics. Monitoring activities were increased substantially both on portfolio and single transaction levels. We focused on permanent dialogue with our clients and discussed tailor-made mid- to long-term viable solutions in cooperation with clients and sponsors rather than going for one-size-fits-all solutions. This is, by the way, one of the reasons why we still abstain from participating in public or private moratoria. On top of it comes the wealth of additional reporting, both to management Board and Supervisory Board and, of course, to our relevant regulatory bodies, be it the ECB, Bundesbank and BaFin. All in all, corona-related client requests or client issues remain on relatively moderate levels. We speak about, presently, 75 to 80 transactions, which make up for about 10% to 12% of our strategic portfolio with requests mostly related to temporary covenant suspensions, changes in covenant structure and postponement of amortizations. Waivers of interest payments have practically no significance or no importance. Of all corona requests, 40% classify as forbearance performing, i.e., all the cases related to changes in amortization schedule or top-up loans or hard covenant changes, while all the other cases, 60%, are remaining simply performing, and that's the majority of which is related to changes in soft covenants. The client requests -- out of this subset, the client requests widely spread -- are widely spread by regions. For instance, Germany covered 20% of the requests, not of the total portfolio, but of the requests; France, 15%, U.S., 10%; and Poland, 5%. And the requests we're talking about mostly focusing on retail. So almost half of that goes to retail, 20% on hotel and 20% on office. We do not make use of the regulatory and accounting easement to retain and report forborne assets if COVID-related. All forborne assets are shown as Stage 2 assets from the first recognition. And as mentioned earlier, we have not suspended stage transfers and have not applied smoothing of risk parameters for the purpose of calibrating Stage 1 and 2 provisionings. Although we registered 2 new corona-related defaults, we added no new provisions to Stage 3 due to the exceedingly low LTVs of these transactions. One U.K. hotel loan added in Q2 '20 was already repaid in fourth quarter. The other, which is a prime asset in New York, serves interest according to schedule against low and COVID-adjusted LTV. In new business, focus remains very much on the prime or A locations, top sponsors, low leverage lending, long-term stable cash flows with focus on tenant quality and lease rollover risk, solid covenant structures as well. Currently, we do not give any new commitments on hotels and shopping centers. Other retail classes, we do only on a highly selective basis with focus on neighborhood shopping and high street retail. Development loans are even more subject to strong risk-mitigating factors or requirements, high level of -- i.e., high levels of pre-letting or pre-sales, upfront equity, increased substantial buffers in long-stop dates in lease contracts and so on. That brings me to Slide 7 with some details on the adjusted valuation process, just to make you familiar with the methodology with the systematics of it. I want to leave that mostly to your reading. Just a few comments on that. While the regular credit review process provides for valuation review on an annual basis, unless there is incident-driven need to look after the transaction, valuation standards in corona times were significantly tightened and are now being performed on a monthly level asset by asset. All new valuations are directly transferred into our risk management systems and become reflected in key risk parameters, such as PDs, LGDs, expected loss and LTVs. While 7% of the total portfolio is somewhat still work in progress, 93% of total portfolio, i.e., of the EUR 27 billion, of all properties were subject to valuation review with specific focus on hotel, retail and office. Logistics and residential, which make up for 20% of that, are without valuation adjustments because not needed. 47% of all assets displayed minor differences, i.e., below 10%. And 24% were subject to a thorough revaluation approach. Slide 9. Key developments can be summarized as follows. Overall investment activity has significantly dropped, i.e., 30%, both in Europe and the United States, even though volumes are still above the levels, which we have seen in global financial crisis in 2008-2009. Furthermore, the take-up levels, both on the rental and the sales side, have overall decreased. Hotel and retail, except food retailing are most significantly affected, while office is largely resilient for now except for relocations, but yields are expected to slightly increase. Logistics and residential assets are stable so far or see even increasing prices. The yields around these 2 types of properties are now at levels which we have seen 5 years ago for prime office, for instance. In all, given there's still high liquidity in the market and the continued low interest rates, we have not seen significant market value reductions all over the place except for individual sectors and markets. Page 10. Now those are the exceptions around well-known asset categories, such as retail and hotel, and a few words on that. It basically follows the same system and the same view and same structure as we did present over the last quarter. So it should not be that much of a surprise to you. Asset class retail was in structural transformation before the crisis. The general trends are unchanged since our last presentation. However, what is to be noted is that the pace of deterioration in some of the subsegments has increased, especially in the U.K. with more retails going under CBA regime, such as Debenhams. U.K. valuation discount and shopping centers since onset of the crisis is around 30% to 50% with some discounts already accounted for before, which we duly reflected in our provisioning exercise. The impact, as we discussed before, they differ by retail class implication. The most affected part, the most affected subsegment, of course, is shopping centers as well as secondary locations and smaller city shopping centers and retail centers in smaller cities, where we would expect a further intensification of the valuation stress. The others, retail parks, discounters, high street properties are very asset -- sub-asset classes where we believe that, in the long run, there will be some remedial development and some restabilization of the property prices. The challenges, which we see on both sides, retail as well as hotels, is that the headwinds are under pressure even in A1 locations and especially for retail insolvencies lead to pressure on rents and allocable costs. The other problem which we see especially in retail, in the subsegments, which I mentioned, that the repositioning options are very limited and depend very much on price. We see some conversions into last-mile distribution centers in some sort of office buildings. But as I said, that's very limited. So our positioning remains, as we communicated, very selective and only investment loans with conservative positioning, average LTV of 51%. On hotels, most of the hotels have been kept open during the second lockdown but at rather low occupancy rates. Now breakeven for business hotels is around 30% to 40% for the inner-city business hotels. But even that breakeven level is not reached in most of the cases over the last quarters. So operators undertake a wide variety of measures to cut cost or to use downtime meaningful, for instance, for renovations or tap into additional capital and liquidity sources. The present increased vaccination activity fosters the expectation of a recovery from late second quarter or third quarter onwards, depending on the way that travel restrictions will be eased and will be lifted. But we also have to acknowledge that hotels dependent on international tourists and business travelers will not substantially recover in short or midterm. It is, of course, visible that leisure hotels, which are focused on domestic guests with good accessibility expected to recover faster. So if you have a nice hotel in the vicinity of Salzburg, they're already fully booked for the summertime. That certainly does not go for business hotel in the middle of Frankfurt. So the challenges are the recovery of the occupancy rates will take time. The other point is liquidity and transactions will be key, and that depends on sponsors as well as on the liquidity, which is kept in the transaction. And it also depends, third point, of course, on the infections and lockdown measures of how they -- how long they will keep on. Again, it's a small portfolio which we have. We continue to focus on prime locations, which we believe mitigates any potential deterioration of property values. Good locations, we believe, will survive or will sell. So in that respect, we'll have limited headache about that particular segment. A few words about office, residential and logistics. As I said, for the better locations, for the prime or A1 locations, we see stable prices, even some yield increases. But overall, we also see rise in vacancies, which are observable. The other point is the bet is still out, the jury is still out on deciding what the structural challenges such as working from home and the hygiene of social distancing will do to the space requirements in office in the long run. Presently, if you talk to -- or if we talk to clients, if we talk to colleagues, present market sentiment is trending to a 20% to 30% structural reduction in space needed due to lasting impact from home office. Now that might be counterbalanced to some degree by higher space requirements because people want to have more meeting rooms, larger meeting rooms and better, larger kitchen -- canteen facilities. But structurally, our feeling is it's trending more towards 20% structural reduction in space, which is coming. And then of course, at the end of the day, we'll have some meaning for the overall price levels and the overall yield levels, which will be seen in markets in the next 1 or 2 years. Residential remains quite resilient. Two trends to be observed. One is, we see some migration from city to country that should have some impact. The other thing is, likewise, the commuter apartments, especially in London and New York, are a bit out of fashion. People do not need an apartment for working 2 days in office and the rest of the week at home. So they rather take the tube or take the train. And we do see a lot of vacancies coming up for that particular submarket segment. But otherwise, overall, residential is a very resilient and stable segment. Logistics. Very popular for investors or with investors. We have seen price increases. There's a limited availability of space -- of usable space in the vicinity of large urban communities, which leads to -- in our views, already to some overheated prices. What is very important what tends to be forgotten these days is that in terms of functionality, there are logistics centers which are multi-purpose and, therefore, transactionable, and there are specific use centers or logistic centers with special cooling unit security or safety standards and so on. I think investors need to be very much aware of that and the implications it has. With that -- with this short coverage of the markets, I turn to Page 12. And in good fashion, I leave the big blocks, NII provisions and cost, to separate basis and will only comment on a few lines on Page 12 alone. The net income from fair value measurement is stable. You may remember that we did account for significant hit in the first quarter 2020 due to corona reasons with credit spread widenings, which left us with minus EUR 17 million on that position. Now that, with the credit spreads gradually catching up until the end of the year, has brought remedy and brought back the levels to what we saw last year. The net income from realizations is, as to be expected, significantly lower year-over-year from EUR 48 million to EUR 26 million, which mainly reflects significantly lower prepayments in the last year, which came down from EUR 4.10 billion to EUR 2.5 billion. Net income from hedge accounting is a positive EUR 4 billion. So there is not much to be reported. Net other operating income stands at EUR 22 million driven by a rather unexpected, I must say, and late release of provisions for tax matters related to earlier tax inspections where we did not expect to get relief from -- but that's what it was, and we had to take it into account. Tax rate of 24% is higher than last year with 17%, as higher risk provisions are not tax-deductible and explains the difference. Page 13 gives you the quick overview on the NII development, which is robust since the last 6, 7, 8 quarters. And it's driven -- continuously driven by lower refinancing costs, where we have profited from good timing from solid credit spreads over the last 2 years. And we, of course, did also profit from some supporting elements from TLTRO III and improved very much from improved floor income, which gave us continued support. And I must say, a big compliment to our origination people who have constantly been able to bring in this harvest. This is -- these positive components are overcompensating for the slightly lower average strategic real estate financing volume; further rundown of the nonstrategic book, in particular, the Value Portfolio; and, as we have it every year, lower contribution from the equity book. Those are the 3 sort of constant factors pushing against further upswing or further uptick in NII. Page 14 shows the development of risk provisioning. As we did indicate with quarter 3 results, we did build a further risk provisioning in Q4. In total, we added up 17 basis points on Real Estate Finance portfolio. Total loan loss reserves almost doubled from EUR 135 million to EUR 261 million in 2020, resulting in a coverage of the fully secured Real Estate Finance portfolio of now 85 basis points. In addition, net additions amounted to EUR 126 million, predominantly driven by EUR 70 million model-based provisions in Stage 1 and 2, of which EUR 28 million belong to the fourth quarter and, as such, reflecting further downward adjustments of economic and real estate forecast given the current developments, i.e., the ongoing second lockdown. As mentioned earlier, we do not apply any smoothing of development of property valuations or GDP developments, neither do we apply any management overlays or relief measures. Total migrations from Stage 1 and -- Stage 1 to 2 amounted to EUR 7.8 billion, of which almost EUR 2 billion were already repaid by the end of the year. Net additions to Stage 3, already commented, of EUR 57 million and solidly related to increases of provisions of assets, which we have already in Stage 3 provisioning since quite some time, which are the 4 U.K. shopping centers. As already mentioned, we had no Stage 3 additions for new cases. Stage 3 coverage ratio now stands at 25%, which -- I may add that -- which is pre-collateral, and if you add the collateral coverage, will come out in excess of 100%. Page 15. We updated our economic scenario assumptions in Q4, in line with the ECB recommendations, taking into forecast -- taking into account forecasts of the central banks weighted with the respective country shares in our portfolio, which is the little bubble which you find in the middle section on the graph. With the portfolio-weighted GDP forecast, over all scenarios of 3.7% for '21, we are positioned conservatively at the lower end of the forecast of the economic institutions, which range between 3.5% and 4.7%. We also revised our assumptions regarding the assumed development of property prices in '21, as market observed visibly lower discounts on real estate than originally expected and feared in 2020, we moved those assumptions by 1 year into '21, thus postponing the originally assumed V-shaped upswing towards '22-'23, resulting in a more conservative calibration of Stage 1 and 2 models, which, I mentioned that, added another EUR 28 million to the Stage 1 and 2 calculation in Q4. So last bit on P&L is the operating cost. Two counterbalancing effect. On one side, personnel expenses are slightly up due to higher FTEs, which is especially driven by IT in-sourcing activity and ramp-up of capacities for regulatory projects and some strategic projects. On the other side, non-personnel expenses were slightly down due to the lower office and admin cost overcompensating for an increase in IT and consulting expenses. We do mention that already. We do hold onto investments in our strategic initiatives and try to compensate those costs related to those initiatives by efficiency improvements. Cost-to-income ratio, I repeat myself, is 42% and thus significantly below market average. Write-downs on nonfinancial assets are mainly driven by scheduled depreciation, almost stable despite the fact that we see a recognition of lease contracts as right-of-use assets according to IFRS 16, since mid of last year. So with that, we turn to new business on Page 18. New business, EUR 7.3 billion, only slightly below our initial guidance of EUR 8 billion to EUR 9 billion despite COVID-related lower investment activities and even more conservative business selection and much lower -- and, of course, much lower than our punching level from 2018-2019, which was around EUR 10 billion to EUR 11 billion. The average LTV stays at 54% on a low level, down from 58% in 2019. And what we do not have and should not have, otherwise, we would have to report it, otherwise, we do not have forced extensions. But what we see is, luckily, significantly higher prolongations with 36% -- making up for 36% of the new business after 20% in 2019. That is good in 2 ways. It contributes to the volume, but it also contributes to higher new business margins as we usually prolong or extend those exposures at a more comfortable margin. No new loan commitments in hotel and retail, as mentioned, apart from some leftovers, which we get from the first quarter. Average gross interest margin is up 25 basis points to now 180, which equates to about 40 to 50 basis points above plan. However, what we also have to acknowledge is the whole market is currently strongly focusing on prime assets and at prime conditions so that we saw some weakening effects already in Q4, and we do expect this phenomenon to persist also into '21. We continue to keep strong focus on Germany with 46%, followed by U.S. and CEE with 12% each; France, 9%; and U.K., 8%. U.K. remains well below our traditional levels when the new business contributed 18% to 20%. But I should also mention here that even after Brexit and in view of the rather -- and despite the unpleasant GDP developments, we still believe that U.K., in particular, London remains a worthwhile market providing good financing opportunities in the future, and thus, we hold on to the proposition there. With regard to property types, the main focus is on office with 50%; logistics, 17%; and residential, 15%, while we've pulled back from the other more critical subsegments. I should add that, although there is a general cautiousness in market, deal pipeline remains surprisingly good in the first quarter and supports solid new business volume for the first quarter at still a very acceptable margin level. Now from new business to portfolio, which is the Page 20 and focuses, as you know, on the sort of risk parameters around the portfolio. LTV, average LTV is 52% and constant since 3 quarters. Despite some revaluations, especially for hotels and retail with a reduction in market values and subsequent higher LTVs, which were compensated by maturities with higher LTVs, which replace -- which were replaced by new business with lower LTVs basically and predominantly explains the relative stability of the portfolio. The LTV levels with almost 50% provide, I think, a solid risk buffer, especially on the background that almost the entire portfolio has been reviewed by now. And lower LTVs, as you know, provide also a strong stimulus for and good motivation of investors to care about their investments and to think about whether they want to give up or not. And that certainly helps discussions when times should be difficult. The improvement in expected loss classification in Q4 is a reflection of repayments from the portfolio. Now next page refers to NPLs. Again, it's a fairly stable picture. It's down by 8% from 410 to -- EUR 510 million to EUR 470 million. The reduction is fully related to restructuring loans, apart from a few marginal technical effects. But it's mainly driven by the repayment, I mentioned that already, of a U.K. hotel loan, which was added to default cases in second quarter '20 and which now is overcompensated or was overcompensated for newly added U.S. loan for high street and retail -- high street office and retail building in Q4, which was triggered by a covenant breach and which is still on our books. Both loans did not and do not require any Stage 3 provisioning due to the sufficient collateral in place, i.e. the loan amounts have been and are well covered by current property value with interest being paid. And as I said, the hotel is now off the books. Now next section is about funding. I think most of the points I have mentioned. It's lower than last year's level, EUR 3.6 billion. Average spreads are very stable despite these forwards which we had in between. The TLTRO III volume, which we participated in, in Q2, amounts to EUR 7.5 billion, of which EUR 1.9 billion were used to replace the former TLTRO II, which allowed also for some funding optimization by substituting other liquid covered funding, i.e. the tentatively more expensive repos. In Q3, we have successfully issued GBP 500 million inaugural SONIA-linked 3 years Mortgage Pfandbrief. It's a long word, first benchmark Pfandbrief since quite some time in a difficult market environment, but with heavy oversubscription, we had EUR 1.7 billion on the table and concluded on EUR 500 million at the end of the day and that with a very attractive spread level at 25 basis points. Meanwhile, spread levels have come down even further back to precrisis levels, which you can see on Page 24 and which also gives you an indication of the good timing which we could employ and which did help us to keep spreads fairly even. Now -- and it was, at the same time, a good preparation for 2 successful benchmark issuances in January, USD 750 million Mortgage Pfandbrief and the EUR 500 million inaugural Senior Preferred Green Bond, both with strong investor demand and issued at attractive levels. Page 25 and -- 24, 25, we basically have covered. So capital. Capital remains solid with 16.1% CET1, which is slightly up over last year based on stable RWAs and slightly increased CET1. I already explained that, that's due to the reduction of the expected loss shortfall. The RWA hovers around EUR 17.7 billion and reflects the recalibration to be expected onto the Basel IV expected levels, which we did in fourth quarter -- third quarter and fourth quarter '19, which in itself, at these higher levels, tends to enhance the stability of the risk-weights. Increases from individual deteriorations due to COVID-19 and new business were compensated by relief from repayments, both in the Value Portfolio and the Public Investment Finance portfolio and by FX and maturity effects. That, in all fairness, we should also say that there might be future -- potential future impacts on risk-weighted assets due to COVID-19-driven reclassifications. Under SREP requirements, nothing has changed. So I will leave that to your reading. I mentioned strategic initiatives at the beginning, and I'm coming back to that now. We did consistently develop and align our business model with current developments, both on the digitalization side as well as on sustainable financing of our real estate business. Our digitalization activities focus on 3 pillars, which you see on Page 29: firstly, the improvement of customer interface on client -- PBB client portal; and secondly, the increase of efficiency in internal clients and credit processes; and third, the development of new income sources via the building of platform CAPVERIANT. Now the first point to say a few words about that. The creation of the new customer portal for commercial real estate clients in Germany, in Europe, in the United States was a major piece of work for the bank in 2020. The customer portal will, ultimately, improve the interface to our customers. Most importantly, we are thereby improving document management and document handling and archiving and are facilitating increased transparency throughout the credit process. Furthermore, this is a first step for workflow digitalization and integration into the bank. From a technical point of view, the portal went live already end of 2020. The real market launch will take place by the end of this month. The next step of digitalization and next logical step is and must be the digitalization of our customer credit process sort of taking over directly from the client portal, what has been deposited there into the processes of the banks. We do start from a good technical basis within the bank, but we want to align the client entry, the portal and our internal processes. We want to make them seamless, and we want to incorporate more flexibility and intelligence by engaging fintech and robot solutions as integral components of the process. Numerous tasks are routine tasks, which have a potential for many for digital solutions, varying from automated data extractions to the use of artificial intelligence to support analysis and decision-making. Where deemed useful, we also coordinate with fintechs or proptechs. There are interesting possibilities, for example, in the area of property valuations. We aim for efficient modular system based on universal workflow providing for increased transparency for the bank as well as for the customers. And we basically put that into an extensive program with a timeline of about 2 years. So it's sizable, significant and it's important not only for the corporation, for the client, but also for the future efficiency of the bank. You can't maintain or you can't afford a 42% or 45% cost-to-income ratio without doing nothing. Third point on this list is CAPVERIANT. Good progress was made during the last year. Besides further improving and developing functionalities, we attracted significantly more users, currently about 400 local authorities and institutional investors, which are linked to the platform, and a cumulative tender volume, which surpasses EUR 3 billion as of now. On the other hand, and in order to further develop and expand the platform, we did look out for potential corporation partners during 2020. And in that context, we had very constructive conversations with Caisse des Dépôts, CDC, French state-owned institution, which is the French KfW, if you want to put it this way. And on December 18, an agreement was signed, under which CDC will, subject to the usual conditions, acquired 28.5% stake in CAPVERIANT. We expect this corporation to create considerable advantage for CAPVERIANT, CDC. Its business line, Banque des Territoires, maintains a very good contact to French public sector clients. This should support CAPVERIANT's market penetration in France and thus promote further growth. PBB will keep a majority stake in CAPVERIANT. It's a strategic investment which we have. And besides the business purpose, which will be developed and further promoted, which is basically to make money at some point in time, CAPVERIANT remains strategically important for the PBB as an important work bench and development laboratory for the further digitalization also of our real estate finance credit process. And altogether, all 3 components, i.e., the customer portal, the digital credit process and the platform, the CAPVERIANT platform, form components of a potential commercial real estate financing platform option in the long run. The second strategic focus, which you find on Page 30, was on sustainable finance. We already launched a project to structurally integrate Green Building criteria into our credit process in 2019. With this project, a set of green criteria has been identified. The focus for 2020 was on the measurement of this criteria, which builds the essential indispensable foundation for sustainable financing. On this basis, we developed a green bond framework, which is in place since Q2 2020, where we successfully issued a first green bond benchmark format in January this year as reported. Currently, we are in the process of developing a green loan concept, which aims to provide green loans to our real estate clients. Furthermore, we integrated ESG factors more strongly and explicitly into our risk management landscape, i.e., into our risk strategy and monitoring and did also address a couple of those points, which you find on Page 30. So summary and outlook. What do we expect for '21? Given the uncertainties in the light of COVID-19 pandemic, it's difficult to provide a concrete outlook. This applies all the more so as we expect economic impact to become more visible only in the second half of this year. In this context, risk provisioning remains the line with the greatest variance. However, all in all, we expect to continue solid operating performance with better PBT compared to 2020 and based on following assumptions: a, stable or slightly higher NII; b, stable operating cost; and three, lower risk provisioning depending on further and future developments of COVID-19 pandemic. The new business volume is expected to be relatively stable around EUR 7 billion to EUR 8 billion. But as remaining focused on prime business with moderate decrease in average gross interest margin, this should lead, all in all, to a moderate growth in strategic Real Estate Finance portfolio. Capitalization is expected to remain at strong levels. And on our dividend policy, we remain on our strategic line, which I mentioned already, 50% regular dividend plus 25% supplementary dividend. That's what we aim for. When ECB has reviewed the overall economic situation in the fourth quarter this year, we will revisit the topic and reassess our options. With that, ladies and gentlemen, thank you very much for your attention. I conclude my presentation. I'm happy to take questions when you have them. Thank you very much.

[Operator Instructions] And so far, we have one set of questions -- I suppose, 2 sets of questions registered by Johannes Thormann from HSBC.

I'm Johannes from HSBC. Three questions, if I may. First of all, on your net interest income outlook, could you be a bit more specific on what margin trends you saw in Q4 in terms of the absolute number and then probably which businesses were still very good and where you saw the biggest pressure as well as for your margin expectations for 2021, where you just said a slight decrease? What levels are you expecting due to what reasons? Secondly, on your other income, it was a chunky EUR 22 million. Could you provide some more details on this? And last, but not least, on the CAPVERIANT-CDC deal, did you make any gain? Was it structured as a capital increase? What do we have to -- how did you set up the capital or the price paid by CDC?

Good. Thank you, Mr. Thormann. CAPVERIANT deal, well, we agreed not to disclose the pricing information, which is probably what you are interested in most. It comes by capital injection in 2 stages. I think it's -- that was the main piece for us. It's been a very fair reflection of the endeavors and the investments, which we made into this digital venture. And I think it's the first time that sort of a bank-led or bank-owned fintech has received an independent pricing take from markets. And I found that very sort of astonishing and very rewarding and did confirm the strategic direction, which we took -- which we decided on for CAPVERIANT to go. Now on other income, a portion of that, I did explain that there are 1 or 2 other factors coming in smaller numbers, collected by the end of the year. But the swing factor is really to the -- related to the tax issue. We did build some reserves for tax inspections, which we had during the year '19 and '20. And as I said, very much to our surprise, the tax man was actually following our arguments, and we did not need those reserves, so we did release them. Nothing we would expect to be repeated. We do have further tax provisions. But unfortunately, I mean, mostly, it's that we pay. Now on the NII outlook, margins outlook, I sort of try to give you a picture around that. If the tide is rising, if there's high flood and you're living on an island, what you do is you see that you're at the highest point of the island when the flood is coming. And that is exactly what is happening in real estate markets just now so that everybody tries to concentrate and to focus on the highest spot of the island, which is where we are, where we have been all the time, but where more friends are sort of joining our stance and our position now, i.e., we have more competition around the same business. And that usually goes along with pressure on the prices, i.e., on the margins. Now that we did see already in fourth quarter 2020, we continue to see that in the first quarter '21. As a general trend, we have been, I think, very good in realizing good margins in 2020. Would we see some levels below that? Yes. Would we think -- would we assume that we sort of roll back to what we had at the beginning of the year? No. Which areas? It is predominantly in the field of those assets, which I tried to explain, office in first locations, in prime locations; residential; logistics. This is where margins have become mostly under pressure.

We have 2 further questions registered from Tobias Lukesch and Mengxian Sun. Tobias, please go ahead.

Yes, 4 questions from my side as well. So first, on hotels. I was just wondering like what kind of scenario, kind of worst-case scenario do you think is or has to happen to see a considerable shift from, let's say, Stage 1 and Stage 2 exposures really into Stage 3? Secondly, on CAPVERIANT, again. How much did you actually invest into that platform? And what is the revenue and the pretax profit that you expect over the next years '21, '23, basically? Thirdly, on RWA, here, my question is twofold. First, in terms of inflation and also revaluation and ready migrations, what did you see basically in Q4? And what do you expect for H1 and H2 '21? And secondly, in terms of the potential expected loss shortfall that we have seen in 2020, do you expect this effect in a similar way to reoccur during 2021? And fourth, on dividends. I mean to kind of compensate for the lost EUR 0.90 per share basically on the 2019 dividend, could you envision to consider a higher payout ratio than the 75% once ECB restrictions are lifted and the crisis is over.

Okay. Lukesch, I try my luck. The dividend piece on '19 dividend and what happens to that, what happens in 2021, we have to see. I can't be very specific on that. We have a good capitalization, but we also have all these uncertainties around the overall economic situation. And as I said this morning, when we had our press conference, I have a lot of understanding for the regulators who stepped in and said you have to keep your money together. You have to keep the capital together because capital stability, bank stability, it's something worth also to shareholders, and shareholders need to respect that and to understand that. And therefore, you know what we are and who we are and how we are. We'd like to talk about that if the picture becomes clear, and we can be more decisive, and we can be more determined on what is coming. And so far, I have to resort to what I said early on. We do maintain our course in the dividend strategy, as we have said it out before. We do, at the same time -- what I say, we do respect very much the shareholders' view, and we'll put that into the scales, and we'll weigh them properly when time has come in the fourth quarter to look at that. If and when ECB should revoke its present policy around dividends, I'm pretty sure there will be a lot of premonition and advice from their side how to keep the money together. But I also -- and that's what I say to the supervisor -- not supervisory side, to the ECB side, shareholders do have a legitimate interest in getting some dividend at some point in time. So I'll leave it at that and turn to the risk-weighted assets and the expected loss shortfall. Now as long as we have built reserves, loan loss reserves, which we set against this expected loss shortfall deduction and as long as the reserves stay there, the shortfall is compensated for and will stay. So that's a rather static exercise. And we should assume that if there's not too much LLPs being consumed in the course of '21 that we will sort of fill the hole and keep the capital levels as we find them now. The RWA rating migrations, so far, we have been lucky in 2 ways [indiscernible]. As you know, we have been lucky in a way that what we have seen as re-ratings and rating increases and increases in risk weights has been compensated either by reductions in volumes on real estate -- sorry, on Public Investment Finance and Value Portfolio or being balanced by other technical effects. I would not rule out that we see further PDs being recalibrated in the course of 2021 with effects on risk-weighted assets as a whole. What you have to keep in mind is that -- and that's something I mentioned earlier on, with the recalibration of risk weights in 2019, we have significantly increased the levels of PDs and LGDs in the model system. And therefore, we have a different degree of stability and variability and susceptibility against worsening of levels than we had before. We are coming from a risk-weighted assets level, say, 2 years or 3 years ago around 25%. We are presently at 50. And if you look at the new business, which we put in even more than 50% risk weight, and that in itself gives, I think, a lot of stability to the bank and the risk-weighted assets and the capital adequacy. CAPVERIANT. That's one of these indiscrete questions, where we try to escape proper answers. I did -- you probably did register that I did give first time some information on the quantitative environment on CAPVERIANT by saying that we have by now 400 interested parties registered on the platform, which is, for a wholesale platform, a good result with more than EUR 3 billion of transactions registered on the platform. We probably -- that's one more I can give on that. We probably will not break even with the whole exercise before the end of '21, '23, -- around '24. That's the time which we will still have to build in for making this venture profitable. We do not calculate against the "profits", i.e., the gains which we have from keeping that as a laboratory for developing our digital ventures, but that's the course it will take. So far, we have managed to build this space, to build this platform without any further stress on our cost line, and I would leave it at that. The last point or the first point on your list on hotels, what kind of scenario shift is required to put that segment more under stress and put that more into the vicinity of Stage 3, it's basically 2 things. One is valuation and the second thing is cash, is liquidity. And first of all, for -- and that sort of speaks for the segmental selection, which we have in our books. When it comes to the top business hotels in inner-city locations, we do not see much of a movement in valuation. So that's why I said early on, these places will either recover or they will sell. And therefore, we tend to be fairly relaxed with the portfolio, which we have. And the other point is who are the sponsors behind that? Are they the big hotel chains? Are they sponsors -- owners to cooperate with them? And are they or both of them willing and prepared to provide liquidity into the transaction to make them survive for the next 1 or 2 years? I think there's very little doubt that within the next, say, 2 years, these well-established, these good places are coming back into business. And as I said, the breakeven level is around 35, 40-ish in terms of occupancy rate. So once they're above that, if they approach 50 to 60, they're almost back to old levels and should be gaining -- and should be, at least, display some profitability. The question is what happens on the journey towards that point in time and who's willing to put money behind that? So -- and we do have cases where we have discussions around that. But as I said, perhaps, it was in one of our earlier calls when we are asked to put our money in, in terms of top-up loan or extension of amortization schedules or whatever, there's a very clear response. We do that if the sponsor, the owner, the principal owner is willing to put more equity in from his side. And that's a quid pro quo to which we stick and which has been quite successful so far. So all in all, for hotel segment, as we have it, it's a matter of timing, it's a matter of valuation and LTV. It's a matter of location. And so far, touch on wood, we are reasonably comfortable with that part of the portfolio which we have.

So to summarize, maybe on the hotel side, so if I make the conclusion that it's very unlikely to see shift into Stage 3 in 2021, is that something you would agree to?

As I said, I mean the whole presentation is about us being careful and cautious. So I should be also cautious on that point. I don't say that it's not going to happen, and we had a case already in second quarter 2020, where we did move one of our exposures into default and Stage 3. But without -- and that's, I think, the important thing, without provisioning needs against that because the initial LTV on that exposure was 40%, 42%, and there was cash in the transaction on account, which would have catered for another almost 2 years liquidity in the transaction. So you may, for, say, more technical reasons or for more tactical reasons vis-à-vis the owner, you may want to put that into default, but you do not necessarily have the need to have Stage 3 provisioning against that. So -- and how much of that we will see -- until the end of the year remains to be seen. It's also a function of the lockdown period, which we are facing.

We still have a small pipeline of questions here with Mengxian Sun from Deutsche Bank and Philipp Häßler from Pareto. Mengxian, the floor is yours.

So 3 questions from my side. The first one is on your office portfolio. You said -- you mentioned that you expect a 20% to 30% reduction in the demand of office space as a result of work-from-home activities. So what is your view on the future development in office buildings? Do you see also a potential price decrease in that space? And the second question is, can you share with us the composition of your nonperforming loans by properties and also by coverage ratio, especially on the U.K. shopping centers? And the last question is, could you also give us an update on the loans which have received a request for changes in covenant structure and temporary payments delayed? Is it possible for you to disclose how big the loan size is?

Now on the last question on loan sizes, I think I gave some indications, but we have it. Yes. No, the -- I think what I mentioned before is that those loans, which are sort of under corona scrutiny, make up for, say, 10% to 12% of the portfolio. And I think that should give you an indication. And I think I also mentioned that 40% of that is under forbearance but performing, and 60% is simply performing. There's no forbearance nonperforming loan under this. So that's, I think, the scope which we disclosed. So forbearance means it's Stage 2 immediately once it is classified as forbearance, but it is forbearance performing, i.e., there are cases of stretching amortization schedules, very little cases for top-up loans, very little -- and there are adjustments to covenant structures, while the 60% I mentioned are requests mostly on waivers for soft covenants, which we find reasonable. Again, for both categories also goes what I said earlier on, on hotels, to the extent that we have been requested to make concessions or to change schedules or to change parameters, we do expect the investors, the sponsors to add on their own money, their own commitments to that. So that's as much as we can say. On the computation of NPLs, I think going back to Page 21, I always thought that we give a lot of information, and you have actually a lot of information, both in terms of regional spread and in terms of distinction between workout and restructuring. Now the majority is restructuring. And the majority in terms of regional setup is on the list below with the U.K. accounting for EUR 348 million, which is more or less precisely the number which is to be attributed to the 4 U.K. shopping centers and which has been relatively stable over time. The other point, which came on top, is the U.S. piece, which I also commented on. So that should be given enough transparency on that side. And the drop, which you have on the U.K. exposure from EUR 396 million to EUR 348 million is due, to a large extent, to the fact that we built EUR 57 million additional Stage 3 provisions, which is being shaved off the value here. Now on office, your first point, the 20% to 30% reduction is a sort of a wholesale indication, which are here in the market. Whether it turns out to be 20% or 15% or 10% or 30% is still to be seen. It's very difficult to prognosticate or to forecast the structural impact, which comes from this actual point. Why is that so? I mean, even if you say -- even if you hear that large firms try to take into account that their employees make use of the possibility in future to work, say, 2 days from home, there is still the demand which also needs to be accounted for, the demand for larger facilities, both in meeting rooms as well as in commonality rooms, kitchens and so on, and that has an offsetting effect. That's the first point. The second point is the question is, how much will that be price relevant now or in future? Most of the firms sit on rental contracts, which are still running 5 years, 10 years to go. So there is a certain lag effect before price adjustments will actually take place. And if they take place, they take place from indirect mechanisms, such as subletting an evacuated space and things like that. That's another sort of number in the equation which needs to be evaluated. And the third one, which is, in my view, in terms of how we look at the strategy and how we position ourselves the most important thing, is the impact on pricing will not be uniformly across the market. As I said earlier, we do observe that prime locations have prime pricings. And everybody, who has a decent office space available in the center of Munich or Paris in the [indiscernible] or whatever or in Central Mayfair locations in London, will have no problems to fill in the space at good prices any time. The price -- the average price will take place and will happen in B locations, in peripheral locations and not so much sought-after locations. And that's why we say, we look at the transaction ability of an asset, and that's why we -- that's where we decide to go or not. So at the end of the day, in 2 or 3 years' time, you will be able to apply sort of an average figure. The discount on prices was 15%. But the truth will be, there will be 0 discount in some locations, and there will be 30% or 40% discount in other places. And the piece of art is to be -- or not to be in the other places. Mengxian, is that at least sort of answering your questions?

Yes. That was very clear.

[Operator Instructions]

Three questions, please. Firstly, on the fair value result. You were mentioning the negative Q1 result of EUR 17 million. We haven't fully recovered this. All those spreads, I think, have narrowed again quite significantly. Maybe you could explain why you haven't fully recovered this or whether you expected to recover in the course of the year. Then secondly, on new business, do I interpret you correctly that the composition of new business, when looking at the regions and the property types will not change in 2021 despite margin pressure in your core areas? And last but not least, on the TLTRO, could you perhaps give us what -- the figure what you expect for 2021?

Well, short and concise, I'll take them in the order you've asked them. The fair value results, we didn't fully recover the effects from credit spread widening. What is left over is sort of a little bit on the interest side. And the third point, which is determining the actual figure, which you see is the negative pull-to-par, which we have in any case. So -- and what is left over is basically the same effect, which we did register in the year before, 2019, which is -- which mostly stems from the negative pull-to-par. So that's on fair value. On new business, I would put it this way. The easy answer is margin versus risk. The composition, and you're right to assume what you said, the composition will not change greatly because exactly for the reasons I tried to make out and make clear. We believe that where we are, this is the right place to be with the prime locations, prime properties and good sponsors. And that risk has a price that we know. To put it the other way around, I'd rather -- I think I mentioned that at the beginning of the presentation, I'd rather forego a loan which we did not make or margin which was smaller than we wanted it to have, than registering a loan on -- booking a loan with a high LLP thereafter. So there's no guarantee in not doing it, but at least we try to keep it this way. And TLTRO, now you know that we draw EUR 7.5 billion on that. We have a 50 basis point advantage. If you split that between '20 and '21, half and half, then you have the figure which you are seeking for.

Well, we have no further questions registered. So I would like to thank you for your continued interest in PBB. We appreciate the time and effort that you put into covering us. You may want to note already in your diaries, May 10 and May 12, which is Q1 results and our AGM. And in the meantime, please keep safe and well and do not hesitate to contact us should you have any questions. Thanks a lot, and take care. Bye.

Thank you. Bye-bye.