Deutsche Pfandbriefbank AG

XETRA:PBB

Earnings Call Transcript 2020-Q2

[Walter Allwicher, Head of Communications]

Good morning, everybody. A warm welcome from Garching. And thank you very much for joining us on this occasion when we want to present our Q2 results. Here with me is Andreas Arndt, our CEO. Andreas will lead you through the presentation that you should receive, which is also online. And after the presentation, Andreas will also be available for your questions. Andreas, please go ahead.

[Andreas Arndt, Chairman of Management Board, CEO & CFO]

Yes. Good morning, and welcome to our analyst call regarding PBB's half year results 2020. I hope this finds you and your families healthy and well since we have spoken last time in May and after some eventful weeks and months in between. After having experienced severe lockdown with restrictions and soaring infection statistics during the second quarter, the overall situation seems to relax somehow despite last increases in infection rates, despite severe regional differences, but in the occasions of hope, such as the latest significant increase in German industrial order index. However, second quarter economic development has also given reason to a significant overhaul of all economic forecasts from all relevant economic research institutes, central banks and government institutions from initially minus 2% to minus 4% to now a range of minus 6% to minus 10%, with the rebound in '21, which, in most cases, does not lead back to 2019 levels. The macroeconomic developments will accelerate structural changes, which also affect commercial real estate. Online will grow faster. Working from home will impact office space requirements and public transportation. Traveling will resume only reluctantly. Retail will change further and faster, apart from the overall macro effects on consumer demand and investments. At the same time, public support measures central bank liquidity, such as TLTRO III and still abundant availability of capital seeking for yield and safe investments seen to counterbalance the negative impact on asset prices, at least to some extent and for some assets. We still see and we still observe transactions at stable prices, although at lower numbers and volumes. Good products always sell, if you want to put it that way. And good products, that means long-term rental contracts, prime properties in prime, prime locations and prime cities. Residential and office are the property types most in demand. So market adjustments will come. They're bound to come. But in our view, they will most likely focus on fringe locations, B properties and other proportionate downside on those places. All that and the unprecedented mixture of developments makes it exceedingly difficult to reliably forecast the 2020 PBT or to deliver meaningful '21 outlook. That being shown, and on a side note, and we've looked at -- and we've analyzed 40 banks, and of which only 2 give guidance for next -- for this year's PBT. That being so, it is even the more so import to remind ourselves and the constituent parts of our bank of the constituent parts of our bank and our strategy. And that is, first, we've positioned ourselves clearly and predominantly in prime or A segment of the commercial real estate, just where the market is now. Second, our business is senior lending. Average weighted portfolio LTV is down 52% -- to 52% despite recent revaluations for retail and hotels. And thirdly, our regional distribution remains conservative, with almost 50% in Germany, 22% in Continental Europe and 10% in U.S. business, of which 75% is in office, residential and logistics across the portfolio. As of now, we consider risk provisioning to be manageable. We built loan loss allowances of more than EUR 200 million over the last 12 months, almost 70% of which being stage 1 and 2 general loan loss reserves. 85% of this year provisioning, i.e., the EUR 70 million is stage 1 and 2. Currently, risk provisioning is not a matter of defaults or not so much a matter of defaults, but a matter of impact from economic forecast on risk provisions of the stage 1 and 2. And fifth point on my list is the operative pillars of our P&L, i.e., NII and general admin expenses, performing well and stable according to plan or better even without TLTRO II taking into consideration. This gives -- for the first half of -- sorry, this goes for the first half of 2020, and this also goes for the entire year 2020. And last point on my list, we're well prefunded into '21 with a 15.8% Basel IV calibrated CET1 ratio, and we are well fortified because of that. We are not immune to pandemics, but we are well prepared. And we stay cautious and conservative but remain confident to deliver a solid operative result in 2020. That brings me to the highlights on Slide 4. I repeat my conclusion. pbb is doing well. Our operating resilience with regards to earnings and costs support further precautionary provisions in Q2. PBT came in at EUR 29 million for Q2 and EUR 31 million in the first half of 2020. NII being very satisfactorily up 5% quarter-over-quarter and 4% year-over-year at EUR 117 million now, driven by lower refinancing cost and floor income on a half year's basis, NII remained stable at EUR 228 million. GAE stayed largely at prior year's level with EUR 49 million, including slightly higher regulatory costs and adding up to EUR 97 million for the first half year. With further risk provisioning of EUR 36 million, we reflect the downward revision of the banks and the market's new economic forecast in Q2, which resulted in EUR 28 million model-based risk positioning in stage 1 and 2. In addition, we had an EUR 8 million addition in stage 3 for a familiar issue, i.e., U.K. shopping center, it's increasing provisions on a known NPL. The new business reached EUR 2.8 billion for the half year. They have EUR 1.1 billion in Q2, not -- admittedly not the largest number. But given the peak of lockdowns in April and May, not too bad, and consciously selective and substantially higher margins. Quarter-over-quarter gross interest margin was up another 15 basis points, which amounts now to 185. And if you take a look back and compare that to the levels we were first quarter 2019 with 125 basis points, that shows the magnitude and the way we have gone. Strategic REF financing volume stays largely stable with EUR 26.7 billion against EUR 26.8 billion as of last quarter. As expected, lower new business volume was counterbalanced by lower prepayments, Public Investment Finance and Value Portfolio are slightly down according to plan and to strategic objectives. As already mentioned last time, prefunding activities result in comfortable levels of liquidity and provides a sufficient buffer well into '21. The majority of the EUR 2.4 billion funded in first half 2020 was collected prior to COVID-19 crisis at very reasonable spread. Further senior preferred funding of EUR 500 million, almost benchmark size, was collected via private placement at, also again, very amenable pricing levels. However, we optimized our funding further with the participation in TLTRO III where we replaced EUR 1.9 billion of the former old TLTRO II and replaced more extensive -- sorry, more expensive repo funding and by the way of doing that generate some positive carry in NII going forward for the second half of 2020. As mentioned, capitalization remains very solid at CET1 ratio of 15.8%, only slightly down over last quarter and on the same as we saw fourth quarter 2019. With regards to dividend, we follow ECB's meanwhile extended recommendation to all SSM banks to defer decision on the payment of the dividends until the end of the year. In line with that strongly worded recommendation, we decided to follow suit and to turn our decision on dividends into first quarter '21. But let me say that already hear now, loud and clear, we stick to our communicated dividend policy of 50% regular and 25% supplementary payout, which we already confirmed earlier this March, inclusive of all qualifications made therein. Again, based on our operating resilience and current risk assessment, we are quite confident to achieve a solid positive full year result even though the impact from COVID-19 will be felt. The ultimate outcome will be definitely depend on risk provisioning requirements and valuation effects, but also important, assuming no further substantial economic and sector-specific deterioration beyond the risk parameter levels cited above. We do not expect any further material effects of risk provisioning stages 1 and 2, but we believe that answer -- we also believe -- also be clear on that, we believe that uncertainty persists with regards to stage 3 risk provisioning, even though we currently have no evidence of any strong increase. So if things stay as they are forecasted now, we should have, by and large, done our homework on stage 1 and 2 German loan loss provisions. So that brings me to the sort of graphical summary or conclusion of what I just said and just point out 3 or 4 things. New business volumes are lower. That's 3 strategic real estate financing, and volumes remain quite stable. I mentioned that operating income from lending business, i.e., NII and NCI as well as cost base are resilient. Cost base is well controlled. Risk provisioning, as I just mentioned, mainly driven by model-based provisioning in stage 1 and 2. And as I said, if we see no further substantial market deterioration and any substantial increase in defaults in the second half of the year, the solid full year results should be achievable. So with that, let me take you to page or to Slide 7 and let's talk a little bit about the COVID-19 impact in the Q2 markets and beyond. While COVID-19 impacts have not yet been visible in the market stats last time when we talked, so April was still quite calm, the picture is changing now. Picture's changing are showing decreased commercial real estate investment volumes in the second quarter 2020 both in Europe as well as in the U.S. as a consequence of the peak of the lockdowns in April and May. Furthermore, we observed slightly increasing office vacancy and declining rents i.e., new office letting take a drop by 1/3 during that quarter, and rental levels are being renegotiated left and right. As mentioned, and that's also important to notice, we do not observe -- we do not yet observe across-the-board reductions of market value as overall liquidity and capital availability in the market is still high. However, with the exception being the U.K. market and especially the U.K. shopping center market, which was already in adjustment mode before the crisis, and the other exception U.S. markets where you see typically a very quick mode to move to adapt to new market valuation. We expect a broader weakening of property prices still to come in fourth quarter 2020 and early '21 as a consequence of the underlying general and structural economic effects becoming more current at that time. Recoveries will only take place beyond 2020 at different speeds in different markets, depending on further lockdown measures, liquidity of the respective market and, of course, especially dependent on retail and tourism. Moreover, COVID-19 will most likely also be the catalyst for structural changes in the real estate sector, especially rising from new hygiene and social distancing standards, digitalization, e-commerce and sustainability matters, which will clearly have far-reaching implications for business we are in. Slide 8. I gave you extensive presentation on those categories on this asset class last time, so please allow me to concentrate on the key messages this time. And I'm sure there will be questions about that. Still, the property classes which are in focus are retail and hotels. Hotels -- most business hotels, you know that's our focus, not -- we are not engaged on leisure hotels. Most business hotels reopened, but occupancy levels are still significantly lower than pre-crisis levels due to hygiene standards, continued travel and events restrictions and the fear of potential regional lockdowns. So it's only a slow recovery. Previous year levels are still not expected prior to 2020, and there's still a long way to go. That's also still business hotels remain our favorite category in terms of hotels. There's a clear trend towards prime hotels with well-known and reliable operators. They rebound more slowly, but more surely, with a more evenly spread occupation, little seasonality and lower breakeven cash flows. Leisure is very much dependent on high capacity utilization as they are more dependent on seasonal guests. Retail, the picture is pretty much unchanged against what we discussed last time, asset class is in structural transformation. The impact differed by retail asset class. Also, that's something which we discussed. The place not to be is out of town as with shopping centers, which are most affected. And the rental levels are decreasing due to shop closures, plus renegotiated trends from existing tenants. But the surprising thing is, and that's very much holds to our portfolio which we have, that ICR interest coverage ratio and debt service coverage, in almost all cases, are sufficient to serve the loans. And the provisions which we accounted for, they're mostly triggered through valuation shortcomings. And we discussed this as well. So overall, lower rent collection rates, but presently, they're coming back. Surprisingly, somewhat surprisingly and also not surprisingly, factory outlets are almost back to pre-crisis levels both in terms of footfall and turnover. So people still believe in fresh air, and people still believe in discount prices. On development, not much to be noted. There's some delays seen in the market, but there's no general standstill. and all operations which we have financed are up and running. United States, as I said, overall COVID pandemic situation is still very much strained, and the investment volumes decreased significantly in Q2. However, I do believe that current job figures give hope that the impact on employment rates is only short term -- is only a short-term effect, and residential markets are not being so strongly affected. Markets expect -- the market is expected to remain attractive for investors. This market values, as I said, adjust swiftly and transparently. And to sum it up on that side, U.S. remains the largest, most transparent and liquid market for commercial real estate worldwide. And it's a very clear, yes, that the U.S. market for us, with all the restrictions which we have to apply, is a bankable and still financeable market going forward. All in all, the situation remains strained that, as I said at the beginning, pbb is well positioned both in terms of risk buffers measured in low LTVs and capitable interest or debt service coverage ratios. And I believe we've done the right thing in selecting the right conservative markets and the right conservative locations. Let me take you to Page 9 on the macroeconomic assumptions. So what does that do to our input parameters or risk assumptions? For the Q1 figures in May, I gave you quite a detailed overview on the assumptions at that time. Meanwhile, GDP forecasts have been significantly revised downwards from leading economic institutions worldwide. As a result, we also significantly adjusted our scenario assumptions in Q2 are in line with the overall sentiment, we revised scenarios significantly downwards and below assumptions made -- which we made at the end of Q1. As you may recall, our base case at the end of 2019 was based still on 1.1% increase of GDP for Germany, which on the way through the first quarter was now revised down to minus 1.9% and now stands at minus 7% for the German market. Overall, for its base scenario, pbb assumes a country weighted GDP reduction of minus 6.8%, which is above the middle between what IMF and ifo say. We supplement these assumptions by an adverse scenario, which goes well beyond the more negative market assumptions of IMF, ECB and OECD. Details of that, by the way, you find on Page 30 in the appendix. While we made significant adjustments to our GDP assumptions, we left our already conservative forecast on property prices almost unchanged and assume a slightly lower recovery. Stronger recovery tendencies will only expected by '22 and '23. The 2020 pre-crisis levels will only partially be reached depending on respective property type. However, also to be noted, this assumes no second wave of infections that requires further measures. With that, let me turn to the financials on Page 11. And I will confine myself to some very general remarks before we jump into the main P&L drivers on the separate slides. As already mentioned, our operating performance apart from COVID-19 pandemic related impacts remain solid with continued strong NII and GAE well under control, with provisioning being mainly driven by model-based provisioning in the light of COVID-19. I come to these points in a minute. The line items to bring to your attention a little bit more detail is the value measurement, which turned slightly positive in Q2 with the EUR 1 million on the positive side after the significant pandemic-related spread widening in Q1. As you probably remember, main fair value position is related to German state, Bundesländer. Only small position to Italian sovereign and to not be named pan-European financial organization. Even though we have seen some further spread widening in the peak of the phase of lockdowns in April, May, spreads came down and came back to end of March levels by end of Q2. And also to put that in order to throw that in, did reduce further in July. Net income from realizations, not very surprisingly, is, on one hand, on previous year's level, having positively benefited from a derivative closing in Q1, which we reported already. Whereas not surprisingly, prepayment volumes and prepayment fees are significantly lower year-on-year. Just to give you an impression, the original plan was that we see EUR 1.5 billion being prepaid first half 2020, out of which EUR 700 million did materialize in the first quarter, but only EUR 200 million in the second quarter. That shows you the direction. Other operating income includes some tax-related provisioning as well as net releases on legal provisions are so by and large, wash. And last but not least, bank levy, unfortunately, due to higher target requirements on the EU level, the demand for another EUR 3 million being put to the reserves. A few more words on Page 12 on the NII. NII remains robust. Robust means 5% up quarter-over-quarter and 4% year-over-year. I think that's quite an achievement these times. The development is especially supported by lower refinancing costs resulting from optimized cash positions held at the Bundesbank and improved conditions of U.S. tender from Bundesbank as well as positive effects from maturities of more costly refinancings. And in addition, we continue to benefit from flows due to the low interest environment. And that was, so to speak, on the positive side. The headwinds were slightly lower average strategic real estate financing volume with roughly EUR 26.9 billion, further rundown of the nonstrategic value portfolio and the low contribution -- and a significantly lower contribution from the equity book. Net income realizations was already mentioned. So that takes us to Page 13. Net income from risk provisioning. Similar to last quarter, risk provisioning is again predominantly driven by model-based additions to stage 1 and 2, taking into account the latest COVID-19-related downwards adjustments in macroeconomic forecast in Q2 and resulting in adjustments to our scenario assumptions, as already explained. In total, net additions in stage 1, 2 amounted to EUR 24 million and in Q2 and EUR 54 million in -- for the entire half year 2020, including -- if we include provisions of balance sheet provisions from lending business, there's some more to look at, so the figure actually rises at EUR 28 million for the quarter and EUR 58 million for the half year. That, together with EUR 8 million addition in Q2 on specific provisioning, amounts to EUR 70 million loan loss provisioning in the first half year, which also means that 85% is model-related. The other way to look at that is, after having added approximately 40 basis points provisions on the real estate funds portfolio over the last 3 years to the balance sheet, we now added another 14 points, bringing the total real estate funds coverage to now 65 basis points with total loan loss in loans of more than EUR 200 million, of which 70% or EUR 137 million is attributable to model-based general loan loss reserves and only 30% or EUR 64 million are specific results. And that -- which I continue to remind you of that, on a fully secured and fully collateralized portfolio with an LTV -- an average LTV of 52%. With that, we have gradually but smoothly increased our risk absorption capacity, assuming no further substantial economic and sector-specific deterioration. We do not expect any further material additions to stages 1 and 2 for the second half. However, and that's the other point, we will keep on saying that, I keep on saying that, while we keep tight leash on every single exposure, uncertainty in stage 3 risk provisioning persists due to the overall volatile and unpredictable macroeconomic situation. For the sake of completeness, one further comment with the economic adjustments made in the light of COVID-19, we transferred EUR 4.1 billion from stage 1 into stage 2 in Q2. But also to put this into some perspective, the bank leaves its pre-crisis accounting standards unchanged, and it allows for stage migration as a result of PD downgrades, which would not have been necessary according to the corona relief measures from EBA and Institute of Auditors. So that brings me to operating costs. There's not much to say. We keep costs under control with general admin expenses slightly up to EUR 49 million in the second quarter and EUR 97 million for the entire half year, which is in line with the plan, slightly above last year's figures. This is due to slightly higher FTE numbers and more prominently dominantly slightly higher cost on regulatory projects. In fact, if that increase will be deducted, we would be back on for last year's figures. The write-down of nonfinancial assets is mainly driven by scheduled depreciation, slight increase of EUR 2 million reflects the recognition of a lease contract as right-of-use assets according to IFRS 16 mid of last year and mainly relates to our new head office in Garching, which you're so very much invited to take a look at and visit us on our new premises, which is still nice even after 1 year. Let me turn to Page 16, new business. As already mentioned, the peak of lockdowns in April, May and the economic implications have led to lower commercial real estate transactions in Q2. One more reason for pbb to act even more selective. That real estate finance new business reached EUR 1.1 billion in Q2 after still good levels of EUR 1.6 billion in Q1. This adds up to EUR 2.7 billion in second half, which is lower than last year, but almost in line with the pro rata -- the old pro rata 22 plan and certainly in line with the circumspect management of a business pipeline in difficult times. Business contracted was -- which we contracted at even more conservative conditions with an LTV of 54%, which is 2% down first quarter and 4 percentage points down versus 2019. And it also reflects a significantly higher gross interest margins. I mentioned that we are presently at above 185 levels with a good way since we started increasing levels since beginning of last year. With regards to regions, we keep on focus in Germany with 47%, while we strongly hold back on the U.K., which accounts for 7% of the new business, and even that is predominantly prolongation business. With regards to property types, continued focus on office with 44%, but significantly lower residential, 5% versus 18% in the portfolio, which has to make that remark, which is not caused by over down-select on that property class, but simply sort of lack of investable opportunities in that particular quarter. And retail and hotel new commitments in Q2, only prolongations of performing loans. As is expected, the overall share of extensions increased to 36%. But also to make that point and to be clear on this, they are all in line with our overall and in light of the pandemic consequences, even increased risk standards, and there are no forced extensions. What about the pipeline? The pipeline is still going strong, surprisingly so. Volume-wise, certainly less than pre-crisis, but a strong eligible choice of transactions with acceptable risk profile. Selection point will be pricing. Of course, risk in the first place. But having said that, selection point will be driving as we have all intentions to see through our adjusted risk pricing and have no intention to let go of flows as part of present pricing. So far, we are holding to our standards and assume the stabilization of our improved margins for second half in 2020 and a somewhat increased new business volume. Segmental reporting, I would leave to questions if you have, because I think most of the things I wanted to say I have said. A few words about the portfolio profile on Slide 19. As mentioned, average LTV is down to 52%, so further improved, reflecting our continued conservative business approach. And the remarkable point is this -- despite some revaluations, which we have already done, especially for hotels and retail, with deduction in market values and subsequent higher LTVs. That was partially offset by maturities with higher LTVs being replaced by new business with lower LTVs, partially by a couple of technical effects. All in all, this provides a solid risk buffer of almost 50% equity in our transactions, just to remind you on that. Expected loss classifications remained more or less stable in the second quarter. And so far, to sum up on that point, so far, the current picture looks quite comfortable. However, figures do not have to point out that as well and do not yet fully reflect the effects from COVID-19 as not every property has been recently revalued. This comes either through specific incidents, of which we hope not to see too many or a general review of sub-portfolios as we have done for hotels and retail or as part of the regular annual review of transactions. As time progresses and as part of the regular review process, we assume a further re-rating of our exposure, resulting in approximately EUR 600 million additional RWA from recalibrations of PDs and LGDs, on top of the recalibrations which we have already undertaken in the second quarter. Page 20, the page on NPLs, which remain on low level, even though slightly up by net EUR 40 million NPLs stayed low with an NPL ratio of 0.9%. The increase mainly reflects the newly added U.K. hotel loan and are triggered by performance covenant breach. However, that's a rarity, I must say. However, no provisioning is required due to sufficient collateral in place. And cash to serve the loan is being trapped in the transaction for a long time to go. And the property value of the loan covers our proceeds in any case, so we consider sale of that transaction in due time. But no further provisioning is required. There were some technical effects, positive FX effects going that I don't need to comment on that. As you know from last quarters, NPLs still include EUR 67 million ECA covered loan, which has been successfully restructured, but which is still in probation period, and Wohlverhaltensphase, in Germany and which we see rolling off third or fourth quarter, if I remember correctly. Workout loans stay marginally low with the unchanged EUR 14 million, and that's I don't know for a long time that EUR 14 million is already there, but it's a long time. And we hope very much and pray for that it stays this way. At this point, please allow me for some further comments on forbearance measures. As already explained, in some cases, customer come to us for different reasons in the context of COVID-19, inquiries are being quite diverse, ranging from extensions of amortization, waivers of covenants and provisions of KfW loans. Most of the requests are related to retail and hotel but not exclusively so. And also several requests related to other properties were addressed. The forbearance measures that we agreed to mainly relate to changes in covenant structures and extension of amortization. We are focusing on finding individual solutions, helping our clients overcome situations, but still adhere to our strict standards. Thus, agreements, in most cases, require and include support elements from sponsor side, i.e., new equity. The number of requests is manageable and still affecting only very minor part of our portfolio. All loans are performing. All other loans, which are under forbearance, which are not performing are anyway covered under NPL, which is a portion of the figure which you see on Page 20 and is in the vicinity of roughly EUR 400 million. So funding. Most of the things being said, strong funding pre-corona crisis and lower funding needs because of lower new business figures for the entire year. In Q2, we still did a EUR 500 million senior unsecured in Private Placements. In total, we collected new funding of EUR 2.4 billion in the first half, a very amenable and relatively steady funding spreads year-over-year. If you look at the funding costs in terms of basis points on our book, that hasn't changed over the last 12 months, which, given the spread movements, which we saw in between, which you can observe in Page 23, I think, is quite an achievement in the smart operation from our treasury people. In addition, we took opportunity to further optimize our funding cost by participating in the new targeted long-term refinancing operations TLTRO III of the ECB with an amount, as already mentioned, EUR 7.5 billion. In this context, we issued another EUR 1.4 billion own use Pfandbrief, which are not included in the figure of EUR 2.4 billion. The proceeds were used to replacing EUR 1.9 billion from TLTRO II and replacing more expensive repo transactions and locking in some positive carry. In total, the financing sector took up EUR 1.3 billion, as you may recall from what you could read in the press, which is 6.5x higher than the participation of the last facility, which was EUR 233 billion. Our drawing compared to EUR 7.5 billion versus EUR 1.9 billion from former issuance, i.e., this is 4x our former allocation. And also to be mentioned, we applied for less than pledgeable asset potential may have allowed for. pbb direct volume is stable at EUR 2.8 billion to slightly increasing this month. Currently, no reason to give any further increase as cheaper sources of funds are available. But I must say, it is still very good to see that we have a stable source of funding, which is also susceptible to price changes and can be quickly scaled. So the overall liquidity remains comfortable, which brings me to Page 23. Secondary market spreads have meanwhile stabilized after a strong widening in March, even though we have ample liquidity into '21, this allows to stay in market on the private placement side. And Page 24, to round up on the funding issues. Pfandbriefe remain the resilient funding source. Market is open and the ability to deliver on funds -- sorry, on Pfandbriefe to ECB keeps cost down. We see a continued strong demand for Private Placements. Retail deposits are established and scalable. And TLTRO provides an attractive source of funding. And U.S. dollar funding possible -- still possible via ECB at attractive rates. What does it mean for the second half 2020? So liquidity-wise, there is no imminent need, but actual funding activities are depending on market conditions. Market -- Private Placements are expected to cover our limited senior preferred demand. Euro-Pfandbriefe are used as collateral for TLTRO, maybe some U.S. dollar issues to match the currencies. And we still consider and contemplate inaugural Green Bond benchmark, which we won't make dependent on the market sentiment. The framework of that, the Green Bond framework is in place. So we are able to and ready to deliver on that field. Capitalization, almost a boring topic, remains strong. We are slightly down from 16.3% to 15.8%, but on the same level which we showed at end of 2019. The SREP requirements are unchanged, so we'll command a buffer against that requirement of more than 70% currently. Let me sum up. Let me come to the conclusions and the outlook, which you find on Page 28. To put it briefly, our operating performance continues to be solid, which is -- which supports further risk provisioning in Q2. NII is robust while we keep operating costs under control. Risk provisioning is predominantly driven by downward revised economic assumptions in the light of COVID-19, mainly based on stage 1 and 2 risk provisioning, stage 3 of less importance, but tightly controlled. And no further negative fair value measurements in Q2. Leaving out the corona-driven provisions, risk provisioning and fair value measurements, PBT for the first half year would have been at EUR 117 million, which is bang on with the figure which we did show first half 2019. On this basis, we are confident to achieve a solid positive full year result 2020, even though significant uncertainties from corona persist and carry on. For the second half of the year, we expect the following developments: new business to increase versus first half 2020 at continued higher margin levels as prepayments are expected to remain low, the strategic real estate financing volume should slightly increase. NII also to slightly increase on full year level versus full year levels 2019 due to positive effects from slightly increasing financing volume, lower refinancing costs and income from ECB's targeted longer-term loan refinancing operations TLTRO III. With the prepayments staying low and the prepayment fees as a consequence also being low. GAE, general admin expenses, is expected to be lower than first half 2020, resulting in slightly lower full year level compared to prior year. And assuming no further significant overall economic deterioration, we expect no substantial additions to risk provisioning in stage 1 and 2 in -- for the second half. Currently, we have no evidence of any strong increase in stage 3, but we do maintain an overall significant -- we do maintain that overall significant uncertainty persists. In general, I would say pbb is conservatively positioned. We have good risk profile with low LTVs and high risk buffers as well as solid capitalization. And based on the current assessment, further anticipated risk provisioning seems to be manageable. We are well prefunded into '21 with 14 -- sorry, with the 15.8% Basel IV calibrated CET1 ratio against an actual risk weighting of 50%, and that gives us, I think, a strong fortification on the capital side. pbb continues to work on cost efficiency and digitalization. As I said, investments in digitalizations are to be continued, 2 main initiatives, which we know -- which we have mentioned and being reiterated here as well. We continue to work on the client portal and expect to go live first quarter '21. And our corporate fintech variant is doing well. We record further increase in number of users as well as significant increase in transactions for the platform. And on dividend, we stick to our principle guidance on dividend and return to you with more news early next year alongside with the forthcoming ECB review. So with all that, we're looking forward to a solid second half 2020. And now I'm looking forward to your questions and a good discussion. Thank you for listening.

[Walter Allwicher, Head of Communications]

Thank you, Andreas. [Operator Instructions] So far, we have Johannes Thormann from HSBC registered and also Mengxian Sun from Deutsche Bank, and Jochen Schmitt from Metzler and Tobias Lukesch from Kepler. And we work them in the order as they arrive, and we'll start with Johannes Thormann from HSBC.

[Johannes Thormann, Global Head of Exchanges and Analyst]

Johannes Thormann, HSBC. Three questions, if I may. First of all, looking at your developments exposure, you actually showed a small increase by -- I don't know if this is just a rounding error from EUR 4.6 billion to EUR 4.7 billion in Q2. Can you, first of all, explain what has driven the increase in this surely more riskier business? And can you probably also share with us how much of this is all pre-let, presale sold? Or is there anything which is really development where it's still already in the fire? Secondly, on your risk provisions. Just some -- each quarter, some single-digit millions in stage 3. And if your disclaimer and current assessment is correct, we should expect still for the next quarter is only single-digit loan loss provisions in total per quarter. Is that a correct assumption? And last but not least, you are guiding for the administrative expenses to decline but said that personnel costs increased due to the headcount increase. Is this still going up versus last year or in the next years? How should we look at that, please? And what is driving the declining admin expenses? Is this just less travel and so on?

[Andreas Arndt, Chairman of Management Board, CEO & CFO]

Okay. I'll start with the [indiscernible] with the G&A. What I said is the total figure for 2020 we expect to be on same level or slightly below what we showed last year. Drivers will be, to a lesser extent, what you have already mentioned, lower expenses on traveling, hospitality and things like that. We haven't been really very lucid and extensive on these things in the past. So there are probably some other firms want more to save than more to spend but that's the figure. The second thing is we have a couple of regulatory projects, which we expect to conclude or at least to scale down somehow by the second half of this year, which should give us relative to the overall cost, which we see some relief. Whereas on the personnel cost side, we would probably expect this to stabilize during the second half of 2020. So those are the imminent components going forward. On the risk provisioning, sort of you try to lead into some assumptions, which I cannot quite confirm, I think. The important thing is, I don't think that we don't have risk provisioning for the second half to come. What I say is that the -- first of all, if nothing changes in terms of economic forecast, i.e., if the overall forecasting stays at minus 7%, minus 8% or whatever and no further other effects will blur the picture, we would not expect to have additional general loan loss provisions going forward for the second -- for the third and the fourth quarter to come. That might still mean that there will be some amounts to be taken care of. Probably the right wording for that is, but not to the tune which we have seen in the first and the second quarter. That does not mean that we may see -- may experience single loan loss provisions for individual exposures. And why do we make that? And why do we make that carve-out and give that point? Given the overall uncertainty, economic uncertainty, and given the fact that we can't really see clearly how this ends up by the end of the year or beginning of next year, we may still see risk provisioning requirements, which we haven't seen in this case, and they might be to be treated individually, and we simply need to take that into consideration. So -- and that's basically the essence of not giving a new firm guidance on PBT. That's the point where I think the jury is still out and is not yet back. On development exposure, we report EAD figures. EAD figures, they move with market with the market. So there is no substantial new amount of business behind that. It's simply valuation-driven, and that's it basically. Otherwise, in terms of risk parameters and pre-let and all that, there's no change over what we have communicated before.

[Walter Allwicher, Head of Communications]

It seems like we've lost some people. And so far -- with Mengxian on from Deutsche Bank. [Operator Instructions]

[Mengxian Sun, Research Analyst]

So 2 questions from my side. The first one is regarding to the LTV decrease? And the second one is a follow-up question on the stage 3 loan. And the first one, regarding to the LTV decrease, you mentioned that the increase of the LTV is due to the valuation -- revaluation of hotel and retail, which is offset by the replacement of new project with low LTV to the higher one. Could you give us some color on the breakdown? How much does the change in LTV comes from the revaluation of the hotel and retail? And then how much is from the replacement? So this is my first question. And the second question is your stage 3 loan takes only a marginal part of the total portfolio, which is less than 1%. Could you share of some of your risk migration assumption under your base and a diverse scenario, i.e., how many stage 1, 2 loans would migrate to stage 3 under the scenario that you've given? And by the end of 2020 and probably going forward into 2021, if it's possible?

[Andreas Arndt, Chairman of Management Board, CEO & CFO]

Okay. The more difficult part is the stage 3 loan migration. I mean it's the nature of stage 3, like But this is event-driven and incident-driven. And I can't really give you a figure on that one here. So we see when it comes. And the LTV side is to be answered as follows. As I mentioned, there are 2 or 3 elements which have influenced the movement of LTVs as much as the movement of risk-weighted assets. One is we see a downward movement of 52% on the LTV from 54% before. This is despite the fact that the revaluation, this sub segmental or sub portfolio revaluation on hotel and retail has taken place. And it's probably easier to answer on an RWA basis, that was sort of EUR 300 million on that side attributable to that element. And sort of goes in line with what I said before. with another EUR 600 million, which we expect from further revaluations to come throughout the rest of the year, but not only on hotels and retail. What we also saw is that's for exposure which did roll off in the course of ordinary business. For ordinary repayments, the LTV was higher than the one for the business which we booked under the headline of new business to be booked. And there was a significant difference in that. And with these 2 elements, that should basically explain the development of the LTV and of risk-weighted assets for the second quarter 2020.

[Walter Allwicher, Head of Communications]

Okay. We have no further questions registered. [Operator Instructions] Further questions registered from [indiscernible] from Barclays.

[Unknown Analyst]

Can we talk about forbearance? So you're saying that you're mostly looking to have individual solutions on this front? But I think you're well aware that there is also the VDP moratorium on amortizations, and I just wanted to understand, in case if there are syndicated loans here involved and with banks that are participating in the moratorium, whether you would also benefit from that in terms of your disclosure regarding forbearance.

[Andreas Arndt, Chairman of Management Board, CEO & CFO]

That's a quick answer. We have no such case.

[Walter Allwicher, Head of Communications]

So we are getting conflicting messages here in terms of further questions. [Operator Instructions] I saw in the meantime, Michael Dunst, but he seems to he disappeared. So if this is a technical issue, please reregister. Next question comes from Tobias Lukesch from Kepler. Tobias, please go ahead.

[Tobias Lukesch, Equity Research Analyst]

A couple of questions from my side as well. Firstly, let me again looking to the kind of stage 3 coverage you show.. So you reported the kind of 65 basis points and only 13% of coverage. How confident are you that with the collateral values, you reach approximately 100% now that you are indicating up to kind of 25% in potential valuation losses on your simulations. I was just wondering why you are so comfortable with that kind of cash covering that you currently have and why you're not working more on this, i.e., indicating potentially a bit more pressure on that side for H2 basically. And also on that, on the risk cost, I just wanted to understand on the GDP forecast that you took into account mainly for stage 1 and stage 2. Is it right to assume that this was only based on the base case and here that the adverse scenario has basically very limited impact on that. And next question would be on the NII development. Thanks for highlighting the impact of the TLTRO. I was just wondering if you -- I mean, EUR 7.5 billion is quite a lot. Could you give us the real net impact of that? So you're replacing kind of EUR 1.9 billion. On the other hand, you also replaced a lot of wholesale funding. So making the simple math of just taking 50 bps on the EUR 7.5 billion for the next quarters is probably not the way we should do it. I mean, you can easily add potentially EUR 10 million per quarter, but how much more do you see actually as a net effect from that kind of redeployment for funding purposes?Again, on risk cost, sorry I missed that. Risk cost-wise, again, is there any figure you can give us as a kind of through-the-cycle risk assessment you have for your portfolio. I'm asking this because a lot of banks are currently giving that, so which gives us a good grasp basically where you see that, that has changed for them over recent quarters. So that would be interesting to see and potentially also a potential peak that you would see in your books.And then finally, coming to capital. You indicated the kind of revaluation will cost you around EUR 600 million RWA, so that's roughly EUR 140 million in capital. Do you expect -- again, what would be the kind of adverse impact that you would see in your model if you were not going for what I assume is just the base case EUR 600 million? And finally, [indiscernible] money effects on the forbearance. Is it possible to give a number here what the actual impact is?

[Andreas Arndt, Chairman of Management Board, CEO & CFO]

Okay. That's a long list, Mr. Lukesch. I'll try to work that down. On the forbearance, I hope I get it right to you. When we talk about forbearance, we talk predominantly about performing forbearance, i.e., cases where we have for limited amount of time in an agreement with the customer to postpone amortizations. We have no -- almost no cases where the borrower owes us interest. So it's current on that side. And where the forbearance is nonperforming, it is included into our NPL portfolio and visible on that side. So -- and the total amount we talk about relative to the total portfolio is very small. So that's what we can say about performance -- sorry, forbearance and forbearance being performing forbearance most in the sense of forbearance which we talk about.Now capital EUR 600 million RWA, I mean we can deliver every sort of dramatic scenario calculation, which gives you much higher figures on the RWA needed in such a case. What we have done is we -- I think we have come to the conclusion that we already show a relative conservative scenario for the GDP assumptions and perhaps even more so a conservative scenario on the assumptions, how commercial real estate prices will develop. And we put that into the system, and we believe that we are scheduling or scaling on a very conservative basis. So is it thinkable that there will be more? Yes, it will be, but I think we're still on the conservative side as that is concerned -- confirmed and in line with the planning which we have.On the risk cost -- through-the-cycle risk assessment. Now precise and concrete figures we do not have, but I think there are 3 aspects which are perhaps important for your assessment. First of all, we apply the GDP assumptions as they come through and not on the through-the-cycle view. So there are quite a number of banks where you find comments in their quarterly reports that, in calibrating or adjusting for GDP changes, they've taken a through-the-cycle view. We have taken a cyclical view and assume that, on average, 6.7% is the figure which we apply and not something, as you perhaps -- when you look on Page 30 of the presentation, not some figures, where is it -- where we have adjusted for in the base scenario of minus 6.8% plus 4.7% and the balance is minus 1.2% as an over-the-cycle figure. That's not been the case.Now that is -- and the second thing, which is also not over the cycle, is that the migrations of which I spoke, the stage 1 and stage 2 migrations, which at least for IFRS purposes we could have stalled. We did not. We did allow for these migrations, and we did allow for the P&L effects coming out of that.So the only thing where we say we have been already very conservative is the assumptions around real estate prices, which you find on Page 31, geographical demonstration, which you know from last quarter, where we give ranges between base and adverse scenarios for various asset quality -- asset categories. And the basic meaning here is that we do assume that up until '21, '22, we would believe that markets still are way below even in a base scenario, way below the original levels. Now that's for '21 -- sorry, for '20, was given as over-the-cycle view in calibration of our systems. So that may give you some hints or some points in terms of degree of conservatism, which we have applied to our risk cost calibration.Now NII impact. What is the effect of the NII or what's the side effects of the TLTRO in terms of NII? I think the most simple calculation is, if you take the EUR 7.5 billion, deduct the EUR 1.9 billion in terms of volume which has been redeemed, we end up with EUR 5.6 billion net. And if you apply 50 basis points net benefit from the transaction, on a half year's basis, you'll arrive at the figure which you can calculate yourself. My calculator shows the EUR 40 million. There might be side effects from replacements of other sources of funding, which are somewhat a little bit more expensive than the one we talked about, but we haven't really classified that or we haven't really separated that. So if you talk about an increase in second half 2020 in NII, that's the effect which we would allocate to TLTRO. The rest is something which is very much dependent on market movements. The -- sort of working backwards on your list, the GDP forecast. Yes, it is predominantly based on the base case. I think we are right to do that because the shift already from minus 1.9% to minus 6.8% was minus 7% and higher discounts in some of the countries is an already conservative move to make. We have set aside to see what may happen in terms of adverse scenario assumptions. We have set aside an adverse calculation scenario calculation, which, to some degree, we have inserted into the overall calibration of risk cost.Predominantly, we base our assumptions on the base case, but there's a conservative element to the adverse case being mixed into that without going into details what the exact mixing component will be.Now on stage 3 coverage, I must confess I'm a little bit lost about your question. I think the 13% is clear how we arrive at that. I assume that I understand your question to that effect and say is 13% sort of enough in terms of way to go when you say you basically covered to 100% through existing collateral which you still have under revaluation? And there I would say, yes, this is the best which we can present at this time. You know that IFRS is not local GAAP, not [indiscernible]. We are not in the position to say, for precautionary reasons, we add another 20% or 30% and make us sort of a sweeping blow building reserves. Where we have built reserves, I think, in the way it is being allowed for and the regulatory required is on stage 1 and 2. And I think if you look at how we have increased loss loans -- loss loan allowances, I apologize -- over the last 12 or 18 months up to a level of EUR 200 million coming from, I think, EUR 92 million or EUR 93 million second quarter 2019. We have built substantial risk provisioning, risk protection for the things to come.But please clarify on your first questions, if I didn't get it right.

[Unknown Analyst]

No, that was all very clear.

[Andreas Arndt, Chairman of Management Board, CEO & CFO]

Okay, thank you.

[Walter Allwicher, Head of Communications]

So we have 2 further questions Michael Dunst from Commerzbank and Jochen Schmitt from Metzler.

[Michael Dunst, Equity Analyst of Banks]

Michael Dunst from Commerzbank. My question is your tax ratio was 26% in the first half of this year. And the question here is, can we take this as a kind of guidance for the full year 2020 because it was in the last years always below 20%? And could you give some reasons for this higher tax ratio?

[Andreas Arndt, Chairman of Management Board, CEO & CFO]

No. As I said, first of all, we do expect some further tax expenses to come. That's been factored in. The other one is deferred taxes and recalibration of that given the outlook for the following years to come. I think it's part of the reflection of being conservative, if we say, we leave it at something like 25% as an indication for 2020 to come. If it turns out better, it's always nice.

[Michael Dunst, Equity Analyst of Banks]

Okay. And it has nothing to do with the fact maybe that you build up high reserves in terms of your loan portfolio? And you're seeing that some of this is not entirely regarded as tax deductible by fiscal authorities?

[Andreas Arndt, Chairman of Management Board, CEO & CFO]

Yes, that's true. Yes. That's a structural element. I should have mentioned that. What we can deduct according to IFRS we may not be able to deduct for tax purpose. That's an important point. Thank you for bringing that up.

[Walter Allwicher, Head of Communications]

One final question comes from Jochen Schmitt from Metzler. Jochen, please go ahead.

[Jochen Schmitt, Research Analyst]

I have 3 questions on your LTV disclosure on Slide 19. First, how much of the assets underlying your loan book were revalued in Q2 '20? Was it just your hotel and retail exposure in total or only part of it? That's my first question.Second, how often do you usually run a valuation process? And third, which average LTV do you expect for year-end 2020?

[Andreas Arndt, Chairman of Management Board, CEO & CFO]

Now the first Schmitt, the first one, how much has been revalued. As I said, we look through each and every item on the hotel and the retail portfolio. So as you can take the figures from the presentation, that's the amount which we look through on a specific basis on a one-by-one basis. Now how often we do that? I would say -- I mean, we have annual reviews. And if we have special places of attention, we do this more often, depending on the sort of warning signals which we have on individual cases and/or sub-portfolios which we think we need to look at.Now a prognostication or guidance on where we think LTV will stand by the end of the year is something which I don't have in store from now.

[Walter Allwicher, Head of Communications]

So we have worked through all of it. That leaves me with saying thank you for joining us today and it was a pleasure. If at all possible, enjoy the rest of your summer. And we'll be back with Q3 results in November, which is still a long way to go. And in the meantime, please stay healthy. Take care.

[Andreas Arndt, Chairman of Management Board, CEO & CFO]

Bye-bye..