Thank you for standing by, welcome to the Millennium BCP nine-month earnings conference call. At this time, all participants are in a listen-only mode. After the speaker presentation, there will be a question-and-answer session. To ask a question during this session, you will need to press star one on your telephone. I must advise that this conference is being recorded today, the 8th of November, 2019. I would now like to hand the conference over to your speaker today, Miguel Maya. Please go ahead.
Good afternoon. This is Miguel Maya speaking. Welcome to the BCP Third Quarter Earnings Conference Call. I will go through the main highlights, followed, as usual, by a more detailed presentation by my colleague, Miguel Bragança, and by our investor relations, Bernardo Collaço. In this quarter, despite continuing under a challenging economical and financial environment, we proceed on track with the goals of our strategic plan, having increased the profitability of our core activity based on a strong and resilient business model, while keeping a firm commitment to improve the quality of the balance sheet. In the international portfolio, the integration of Eurobank in Poland is running smoothly and according to our projections. The legal merger with Bank Millennium was an important milestone, successfully concluded on the first of October, and we expect a fully operational integration during the current quarter.
Still in Poland, we are closely monitoring the implications of the European Court of Justice ruling related to FX mortgage loan of a specific local bank. Although this ruling cannot be translated to Bank Millennium's loans, which have different clauses and practices. Lots of things have already been said by different entities about the possible implications. Public comments were made by rating agencies, analysts, the Polish Bank Association, and so on. The only things that are crystal clear for us is that our loan contracts are different from the one judged, and the decisions will be taken case by case by the local courts under the full extent of the Polish legal framework. That we are and will continue to be fully committed in giving this matter the attention it deserves, and we will be, as always, transparent to the markets.
Starting on page 5, as you can see, there was a year-on-year profitability increase of 5% on the third quarter, driven by the performance of the activity in Portugal. Net incomes stood at EUR 270 million. The activity in Portugal contributed with more than EUR 125 million, representing a growth of 7% comparing with the same period of last year. Net income in Poland, excluding Eurobank integration costs, grew by 7% year-on-year. This improvement of profitability in consolidated terms was supported by a 7% growth in the core income, mainly driven by a 9.5% increase of the NII, confirming the quality of our franchise at the several geographies where we operate. The consistent work performed both in improving the quality of the assets and in managing the loan book is driven the continuous decrease in the cost of risk, led by a decrease of 12% in the impairment.
Moving to page 6, I would like to highlight our strong commitment and determination in keeping the pace of reduction in NPEs, particularly in Portugal, where we have already reduced EUR 1.1 billion this year, with almost EUR 400 million in the last quarter. Over the last year, we have reduced NPEs by EUR 1.7 billion in consolidated terms, reaching a stock of EUR 4.6 billion in September, including Eurobank. This reduction was achieved alongside the reinforcement of the impairment coverage to 55%, while keeping the total coverage stable at around 107%. The cost of risk continues to converge to the strategic goal of 50 basis points we set for 2021. If we exclude the first day impairment of the Eurobank acquisition, the cost of risk in September was 68 basis points.
On page 7, we can see that throughout 2019 we have confirmed our capacity to organically generate capital with the Common Equity Tier 1 standing at 12.3% in September. This organic generation of capital more than absorbed the expected impacts from Eurobank acquisition and also the impacts on the pension fund from the adjustment in discount rate and this reinforcement after the agreement with unions regarding the wage update. In September, we completed another stage in our strategy to optimize the capital structure and the funding sources, having issued an additional amount of EUR 450 million in medium-term subordinated notes that qualify as Tier 2 own funds. The swift execution of this new issue reflected, once again, the market's confidence in Millennium bcp. The total capital ratio reached 15.7% in the third quarter.
We remain clearly above the SREP requirement for both ratios, having a capital buffer of nearly EUR 1.2 billion. Our business model is driving the bank's capacity to consistently improve the business volumes, increasing the trust that customers place in Millennium bcp. The loans to customers increased EUR 3.5 billion year-on-year, while the performing loans have increased by EUR 5.2 billion in the same period. Moving on to page 8, we confirm that Millennium bcp strong franchise and market recognition is also being reflected in our capability to increase the customer base. Excluding Eurobank integration, we added 246,000 customers globally, more than half of which, 141,000 in Portugal. This growth of the customer base is consistent with the strategic plan we set until 2021. We are implementing a mobile-centric transformation of the customer experience to enhance our commercial and relationship banking business model while obtaining productivity gains that boost the bank's efficiency.
The number of mobile customers is increasing at a faster pace. Year-over-year, the number of mobile customers grew by 28% overall, and that growth was 37% in Portugal. On page 9, we can see that the customers are engaging with the bank's transition towards digitization and an increasingly mobile-centric approach. This transition goes beyond the front end, involving the transformation of the back office operations and processes. We are integrating technology in our operations and optimizing processes to improve efficiency, obtaining productivity gains while providing customized and increased value service to the customers. Let me highlight that with the introduction in 2019 of robots in our operations department, nowadays, more than 5,000 operations a day are being performed by these technologies. In the first nine months of 2019, the number of new customers using our mobile apps exceeded the increase in 2018 full year.
As of September 2019, digital customers already represent 58% of the global customer base, and 39% of our customers are regular users of the bank's mobile services. On page 10, we can see the effect of the launch of our new app in Portugal in May, which has increased and enhanced usage of our mobile services. The superior user experience, the increased functionalities, and convenience of the new app have proven to be correct interpretations of the customer's expectations for their daily interactions with the bank. Our strategic and business model has been validated by credit rating agencies, as shown by several recent rating upgrades. Since June, we are investment grade for senior debt from DBRS, and in July, Moody's also issued the same qualification for deposits. More recently, in October, both S&P and Fitch reviewed their outlook for BCP from stable to positive.
Before concluding, I would like to emphasize that our retail business model has proven to be resilient, and our relationship with corporate and SME customers is benefiting from the fact that we are the sole commercial private bank that consolidates in Portugal. On top of that, our profitability will continue to benefit from a significant and consistent reduction in the cost of risk. I now give the floor to my colleague, Miguel Bragança.
Good afternoon, ladies and gentlemen. It is once again with great pleasure that I announce here our Q3 results. As you see on page 13, in spite of the challenging environment in terms of interest rates, we continue to grow in terms of core income, being core income, net interest income, and commissions presenting a 7% growth. These after costs translates into a core earnings growth of around 5%. After non-usual items and other income, including trading gains, this means that our pre-provisioning profit is still growing in spite of the challenging environment. With the normal trends towards a more reasonable cost of risk, we were able to reduce the impairments on the 12%, and this together with a sharp decrease in NPEs that we have shown, so that our net income before tax grows 14.4%.
As you have seen in Q2, given the new framework for tax deductibility of impairments and the reduction in the Euribor, we have to write off a part of the DTAs that we had in our balance sheet. This explains this small de-recognition of DTAs. This explains the higher tax rate already in Q2. Nevertheless, we were able to present relatively to last year, a growth of 5% in terms of net income. Let's now see it in detail how this translates in different operations. In page 14, you see very clearly that the intrinsic profitability of the bank has increased EUR 54 million, i.e., 20%. We have here a lot of issues that are not so manageable, so to say, including the gains on Portuguese government debt that are not so recurrent and the increase in mandatory contributions.
This 20% is clearly aligned with our objectives for this year. In terms of net interest income, we are able, as you see on page 15, to preserve our NIM. That stays at 2.2%. This sustainability of the NIM is composed of a very reversed, I would say, performance in terms of NII in Portugal, that in spite of the decrease in the Euribor, we are able to maintain a NIM of around 1.7%, going down by less than 10 basis points. The growth in the international operations of 3.2%. This 3.2% is linked to this 21% growth in terms of NII. Of this 21% growth, roughly half of it has to do with the acquisition of Eurobank and the other half, as you may see in the graph, is real organic growth. In terms of fees and commissions, this is a tale of two parts.
We have, on one hand, everything that is market related, more investment banking and less recurrent, going down 15% for reasons that you know well. A part of this has to do with asset management and with the risk appetite of the private clients. The more stable commissions that have to do with our customer franchise are going up 5.5%, and this 5.5% very well balanced between the Portuguese operations and international operations. As you see, 5.3% in Portuguese operations and 6.1% in international operations. Here again, in international operations, roughly half of the growth is linked to Eurobank acquisition, and the other half is linked to the intrinsic organic and client growth. In terms of other income, comparing with the accumulated other income of the third quarter of last year, we see here a small reduction. It's quite slight, but a small reduction.
The responsible for this small reduction is the one-off effect that we had presented already in Q2. We don't have really any one-off here in Q3 that had to do with the reduction of interest rate and its impact on the insurance activity. As you may see, in terms of Portugal, you see that we are able, in spite of that, to maintain stability in terms of other income. The value this year is broadly equal to last year in terms of Portugal. In terms of the international operations, the main responsibility of the decrease is linked to the mandatory contributions, so contributions to bank tax and resolution funds in international operations. That adjusted for this, we have a growth, as you may see on page 17. In page 18, we see here the costs. The costs deserves to be explained.
You see our recurrent costs in Portugal are growing 2.9%. This 2.9%, a lot of it has to do with the transformation that we are doing in the bank. I would like to highlight that even the low cost growth that we are able to achieve in spite of this transformation, as you do the digital transformation, typically has a J-curve effect where at the first moment there is an increase in cost to reap the benefits later on. We have been able to transform the bank, as Miguel Maya has just shown, with a relatively minor increase in terms of the recurrent costs in Portugal. In terms of the international operations, you see also that there is a growth in terms of the recurrent costs of around 18.3%. Here again, roughly half of it has to do with the Eurobank acquisition.
On a pro forma basis, we will be growing half of it, and a large part of this is linked also to the labor market in Poland. That, of course, has the other side of the coin, that is the top line in Poland is also growing very well. As you see on page 19, this evolution of both of the top line and of the costs is assuring that you maintain a leading position in terms of cost to core income. We are leaders, as you see here in Portugal, and we compare very well with the averages in the Eurozone in spite of this low interest rate environment. We have already changed our model to a low interest rate environment. We have more than 1.2 million customers that pay regular, transparent fees, as I say it very openly.
We want to distinguish ourselves through service quality, through client loyalty, but clients that also see enough value in us and in our services that are predisposed to pay for this value in a very transparent way, typically through paid accounts and through monthly fees. As I say, somewhat, we are the Netflix or the Spotify of banking in Portugal. A very transparent fee structure, but also a very modern business model. The impairment and provision charges came down as expected, excluding the additional day one impairment that don't have an impact in terms of capital, in terms of the acquisition of Eurobank. Our consolidated cost of risk is already at 68 basis points. This has been achieved mainly through the reduction of the cost of risk in Portugal that you see here has come down from 102 basis points to 74 basis points, so around one quarter.
This is a trend that may have some oscillation, but broadly, we will expect to continue to achieve our target of 50 basis points by 2021. In terms of NPEs, this continues to be a strong focus of the bank, as you see. We reduced the 27% of our NPEs, broadly aligned with what we have commented of a reduction of around 25% a year in terms of NPEs. Actually, on top of this, our NPL ratio, so NPL 90 days, so really past due ratio, is already below 5%. We already have a ratio of 4.7%. Our EBA ratio, that includes securities and off-balance sheet exposures, is at 5.9%. The NPL ratio, including unlikely to pay is at 8.4%, coming down from 12%. A very sharp reduction of around one third, as you see. This sharp reduction has been mainly in Portugal.
That is most of our operations. Eurobank, in spite of contributing positively to the ratio, of course, in terms of absolute values, it brings also some NPE. It is a low NPE bank, but nevertheless, it explains this small increase in the stock of NPEs in the international operations. I will pass the floor now to our chief IR. I'm sorry. No. Okay. I'm sorry. I'll continue here. I'm sorry. Just here. In terms of the net loans to deposits ratio, in terms of the capital ratio, I'm sorry, page 27. I'm sorry for this small disturbance. We see here that our capital ratio continues to evolve positively from 12% in December to 12.3% right now, broadly aligned with our target of 12%.
If you see the total capital ratio, it benefited from the issuance of the AT1 and of the subordinated debt, reaching a level of almost 16%, which is a very good level for a bank that is focused on retail operations such as us. The leverage ratio is a very good leverage ratio. As you see, we are a bank that has a quite high risk-weighted asset density. This means that if we have a good capital ratio, we have an even better leverage ratio. In terms of the activities now per regions, I'll pass now to Bernardo Collaço.
Okay. Good afternoon, ladies and gentlemen. Let's start on page 30, looking at the Portuguese operation where net income on the first nine months increased by 7.1% to a level of more than EUR 125 million, driven by the stability on top line, that means NII and commissions, and a significant reduction on the credit loss charges with more than 28%. Banking incomes stood close to EUR 1 billion, showing the resilience of the business model in a more challenging environment for the banking sector. Recurrent costs increased by 2.9%, but bearing in mind that we have and we are doing some investments on the business model and recruiting young people, and in terms of non-recurrent costs, reached EUR 24.4 billion, split by EUR 12 million of restructuring costs and EUR 12.4 million related with the compensation for temporary salary cuts explained on the last quarter.
On page 31, it is shown NII evolution on this low interest rate environment. As you can see, there are some positive and negative impacts that's worth mentioning, and that explain the increase of EUR 4.3 million to a level of EUR 600 million year to date. This evolution is driven by positive effects of EUR 21.5 million coming EUR 3.6 million from the expansion of the performing book, EUR 7.2 million from the decrease of the remuneration of time deposits, and EUR 12.7 million from the lower wholesale funding cost. On the negative side, contribution was negative in EUR 17.2 million, driven by the decrease on the average loan spreads reflecting the normalization of the macroeconomic environment and EUR 7.3 million regarding lower yields on security portfolios. On page 32, when we look at the deposit spreads, back book is at a level of 55 basis points.
That means an improvement from last year's spread of 57. We have also to consider the decrease on the average three-month Euribor that happened on these related periods. All in all, there is a reduction of four basis points. There is still some space to capture additional value because front book is being priced at an average of 40 basis points in the first nine months of 2019. Regarding the spread on the performing loan book, it has been stable at 2.72% and NIM stood at 1.7%, nine basis points lower than last year, which reflects, as mentioned before, the normalization of the macro environment. On page 33, regarding commissions in Portugal, we can see an increase of 1.2% to EUR 357 million. Regarding banking fees and commissions, we went up 5.3%, mainly driven by fees related with loans, bank insurance, and customer accounts.
On the other hand, market-related fees offset a better income in this line and decrease year-over-year by EUR 11.8 million. Other income, as already explained by Mr. Bragança, decrease was mainly due to the lower equity earnings with the impact from Ageas, the insurance company, registered on the second quarter, as well as the lower contribution from other stakes that BCP has. Other operating income reflects the mandatory contributions that were stable from last year and the positive impact from sales on other assets. On page 34, regarding costs, we have to consider, as mentioned before, some one-offs related with restructuring, around EUR 12 million, and the compensation for the temporary salary cuts paid last quarter to employees with an amount of EUR 12.4 million. Recurring costs, with a slight increase as mentioned as well, with EUR 2.9.
In terms of net employees, we increased by 1.8%, but we are still reducing part of the workforce by early retirement and mutual agreements, although, as it was mentioned, we are adapting the structure for people more related with the digital and so on, some younger employees. Moving to page 35, in terms of asset quality, we are continuing the downward trend of reducing the NPEs. Year- on- year, we were able to reduce EUR 1.9 billion that reflects a 33.4% reduction, reaching a level below EUR 3.7 billion. At the same time, we continue to reduce the cost of risk to 74 basis points, whereas it was above 100 basis points in September 2018. Over the last 12 months, NPE decrease was done through a combination of EUR 670 million net exits, EUR 534 million write-offs, and EUR 700 million of sales.
The performance year to date has also been very positive, with a decrease of EUR 1.1 billion, again, with a balanced combination of one-third of net exits, one-third of write-offs, and one-third of sales. On page 36, NPE coverage show that top total coverage is still above 100% for both individuals and companies, and for both NPE categories. It means 90-day past due and other NPEs. NPE coverage by provisions is at 54% for the whole NPE book, but stronger for loans to companies, and in particular, for the 90 days past due segment. On page 37 and regarding foreclosed assets and restructuring funds, there was an important decrease of 27% on foreclosed assets from last year, with net value in September 2019 at a level below EUR 1 billion.
We increased sales, as you see, by more than 25% in terms of numbers of properties, and the bank continues to register some profits on those transactions. Corporate restructuring funds decreased 5.3% to EUR 969 million at September 2019. That still shows a decline trend, although at a lower pace. On page 38, moving to volumes in Portugal, customers funds were up 5.7%, driven mainly by the growth of individual funds, and performing loans increased also by 4.5% with more than EUR 1.4 billion loans that were unable to compensate the large sale mentioned before, on the reduction of the NPEs by EUR 1.9 billion. If we look on page 39 in more detail on the performing book, increase of 4.5% already mentioned was mainly explained by the strong performance that we have in providing loans to Portuguese companies.
That represents almost 50% of this annual growth and reinforce Millennium bcp role as the bank for Portuguese companies. Looking at international operations and moving forward to page 41. As you can see, contribution from different operations stood at a total level of EUR 131 million on the first nine months of 2019. This compares with EUR 141 million of the first nine months of 2018, or if you want, less 6.6% versus last year, first nine months. If we're not considering the integration cost of Eurobank, contribution from Poland, instead of being minus 2.6%, it stood at plus 7%. The slightly lower contribution from Angola is explained by the amortization of the effect of IAS 29, and in Mozambique, contribution has been stable at a level of around EUR 75 million. If we look at Poland on page 42, net income decreased 2.6% to EUR 124 million.
As mentioned before, if you don't consider the integration cost of Eurobank, net income had increased by 7%. Return on equity is above 9%, and common equity run at a level of 17.1%. Net operating revenue increased almost 25%, considering the accretive value of Eurobank since July, as well as the natural growth of Bank Millennium. Operating costs were impacted by a higher contribution to the resolution fund by increased staff and integration costs of Eurobank. Cost to income without those effects is relatively stable at a level of 46%. On page 33, some highlights about the integration of Eurobank. That proceeds according to the plan. Legal merger is complete in October. The first full merger, that means a single brand and IT system, to be complete this month.
Another point that is worth to mention is non-usual costs related to the integration have a total of EUR 35.5 million in the first nine months of 2019. That I explain EUR 14.9 million of integration costs and EUR 20.6 million of additional impairment, but most part of it at the first day impairment that we mentioned on the previous quarter. On page 44, NII increased by more than 30%. That reflects the consolidation of the Eurobank business organic growth of Bank Millennium, and higher loan volumes driven by cash loans that is also responsible for NIM increase from 2.6%-2.8%. Commissions and other income were up 13.5%, and if we don't consider tax on assets and contribution to mandatory funds. Costs were higher, but if we exclude, once again, Eurobank natural costs and integration costs, Bank Millennium costs were higher only 8.5%, reflecting some IT investments and wage inflation.
Bear in mind that the Poland GDP is growing fast at a level of around 4% and the unemployment rate is very low. Moving to page 45, in terms of asset quality, you can see on top left chart that NPL ratio has decreased, but with a stable level of around 2.6%. With a cover ratio at more than 100% at 107%, and cost of risk at 80 basis points. If we exclude the additional impairments related with Eurobank acquisition, costs of risk should be at a level close to 61 basis points. The cost of risk is naturally higher than before due to the fact, and excluding the extraordinary events, you have to consider the change of the loan portfolio structure.
As you know, we have a bigger share of cash loans as well as total loans on our portfolio, which, of course, drives up the cost of risk. Regarding volumes on page 46, customer funds increased 28.6% with a contribution of around EUR 2 billion from Eurobank, and loans increased from EUR 12 billion- EUR 16.3 billion, or plus 35%, driven by the contribution of around EUR 3 billion from Eurobank, and the natural increase in new production that was seen over the past at Bank Millennium on cash loans and mortgage loans. On page 47, moving to Mozambique, net income stable at a level of EUR 75 million with a return on equity above 20% and a capital ratio close to 48%. Net operating income grew 2.3% to EUR 187 million, and costs increased 6.4%, but still with a cost to income below 40%.
On page 48, regarding, again, Mozambique, NII has been stable at a level of EUR 140 million and NIM is still at a level close to 10%. Commissions and other income were up 11% and operating costs up by 6%. Moving to page 49, the NPL stock in Mozambique is going down from last year, and the NPL ratio is higher because of the decrease of the loan book. Loan loss reserves went up to 67% from 60% registered at the same period of 2018, and cost of risk with a decreased trend reaching a level of 300 basis points from 340, and in line with our conservative strategy. Regarding volumes in Mozambique on page 50, customer funds were up 4.3%, and it is also important to see the increase on demand deposits and the decrease on term deposits that allow Millennium bim to reduce the cost of deposits.
Loans to customers decreased 19%, reflecting a conservative approach under the current challenge environment that is in Mozambique. Thanks for your attention. Before we move to Q&A, I will turn again to Mr. Miguel Bragança for some final remarks.
Thank you very much, ladies and gentlemen. As you see here on page 52, we are on track for our 2021 ambition that we have presented to the market. This ambition has several levers based on the transformation of the bank. We are growing our active customer base, and we are converting these active customer base in a much more mobile and digital customer base, which will enable us, over time, to reduce the cost of servicing with quality this customer base and achieving our planned ROE targets. This within the limitations and within the self-imposed restrictions of having a very solid capital ratio and of having a very solid balance sheet from a liquidity standpoint.
In terms of the NPE stock, I would say that we are clearly overachieving or anticipating, I would say, our objectives as we have just presented, and we are doing so in the context also of the reduction of the cost of risk that will be, at least for the Portuguese operations, a key lever for achieving our targets. I will open now the floor for question and answers. Thank you very much.
Ladies and gentlemen, we will now begin the question and answer session. As a reminder, if you wish to ask a question, please press star one on your telephone and wait for your name to be announced. Please stand by while we compile the Q&A queue. If you wish to cancel your request, please press the hash key. Your first question comes from the line of Sofie Peterzens from JP Morgan. Please ask your question.
Hi, here is Sofie from JP Morgan. I would have two questions. My first would be on the EUR 3.3 billion of FX mortgages you have in Poland. Could you just give a little bit more details, how these contracts that you have in Poland are different to your competitor, and what do you expect in terms of losses? I know you gave a little bit of details in the beginning of the call, but if you could give a little bit more flavor and how you think about the mortgages going forward. My second question would be on capital. What kind of capital headwinds should we think about going forward, including any changes to the pension discount rate?
TRIM, I know you have guided that it will be a small impact, but if you could give us an update here and anything on EBA guidelines and Basel IV. That would be my questions. Thank you.
Sofie, thank you very much for your questions. In terms of the FX mortgages, as you know, this is an old issue. We ceased to originate FX mortgages in end of 2008, so more than 10 years ago. This has been more or less a recurrent issue in the Polish market. At first, the main risk that was seen related to these FX mortgages was a legal change as it happened in other countries of this zone. Even in the middle of this year, there were some rumors that the Polish parliament could impose a legal change. Fortunately for us, those legal changes that would be immediate and directly applicable to all mortgages was not imposed. In the meantime, this trend or this difference, it translated more to the courts.
Effectively, what happened is that by using a European directive from 1993, so a very old directive that could already have been used, a Polish court asked some questions to the European Court of Justice, trying to clarify some points in terms of the abusiveness or the consequences of the potential abusiveness of these contracts. As you know, we are not party in this context. Basically, this is between a client of another bank and another bank. Basically, what the European Court of Justice says is the following: If these clauses were to be deemed abusive, we are not speaking about the indexation clause, but about the FX spread clause, so the clause, by which banks make money in the bid and ask of an FX deal.
If the FX clause is deemed abusive, this in principle should somehow be seen together with the indexation clause, which is somehow bad news. In principle, in this type of situation, if and only if it is abusive, this would in principle generate the annulment of the contract. I'm simplifying it a little bit. If you need a legal opinion, please contact your lawyers. Just to make this simpler for everybody, this was the first opinion from the Polish Court of Justice. Effectively, this is an opinion of the European Court of Justice to a Polish court, the Polish court has not even decided this case. This was an opinion that was given on the 3rd of October of this year, not very long ago, a couple of weeks ago.
The final decision of the first instance Polish court will only occur next year and probably even this specific case, I would think it will not be decided on final terms during next year. I just want to highlight this point. It's a very early days. What we can tell you is the following. This issue around the litigation regarding the abusiveness of the case is also an old issue. In the last three years, by the information that we have, the banks have won, in terms of this issue of indexation, around 90% of the cases. Okay. We have not lost one single case in terms of indexation in final form. Our cases that achieved the final stage, we have not lost one single case.
All the cases that achieved the final stage were won by us, in terms of the system, the information that we have, the system has won more than 90% or around 90% of the cases in the last three years. There is not a lot of new information regarding what are the decisions of the courts after the 3rd of October. Based on some press information that is typically more biased towards clients, what we see is that the decisions are all around. There are some decisions that continue to give reason to the banks. There are some decisions that somehow put in question or challenge the effectiveness of the indexation clause. There is not any final conclusion on this matter. What I can tell is that we have absolutely no decision from second instance or above since the 3rd of October.
In our case, we have absolutely no new information from second instance or above. From first instance, the decisions are very little and all around. Even yesterday, we won a case. We won a case in first instance, and won totally. The decisions are all around. It's very early days to give any view on this issue. What I can tell you is the following. Based on the European Court of Justice decision, first one has to see whether the clause is abusive or whether the clause is not abusive. Okay? To see whether the clause is abusive, we understand it's very important to look at three things. The first thing is whether the client has an alternative or not. In our specific case, the client had always an alternative since the beginning of paying in foreign currency.
Even if our bid-ask-spread were too high, which it wasn't, the client could always pay the loan in foreign currency, which makes a big difference. We don't know the details of the specific case that was in Europe or in the Polish courts, but we know that some of the banks don't have this, at least since the beginning. This makes a lot of difference. A very important situation also is the way that the clause is worded and whether it is transparent, whether it is easy to understand, and what type of client you have on the other side. Whether the sales process is correct or whether it isn't correct. This changes on a bank-by-bank basis, and we are quite comfortable in terms of our sales process. A third issue that we cannot forget is the following.
In terms of the European consumer law, the European consumer law basically only classifies as abusive clauses that are simultaneously unbalanced and that are not essential clauses of the contract. The clauses have not to be considered essential clauses of the contract, in general terms, to be considered abusive, unless they are worded in a very untransparent way. What we think is the following. If one bundles together both the spread and the indexation, it becomes so important for the contract that it becomes an essential part of the contract. Becoming an essential part of the contract makes it much more difficult to be classified as an abusive clause, according to Article 4 of the directive. It's very early days. We know our process. We feel very comfortable with our sales process. Of course, we have to follow it closely. We are not being complacent.
It was on the 3rd of October, and since the 3rd of October, we have, in our specific case, no new information, and even some information are positive, I would say. Let's wait, let's see until we have more data than to form any type of conclusion. In terms of capital. Going forward, of course, our bank and our group generates capital on a pre-provisioning basis or a very strong capital on a pre-provisioning basis. I think this is important. The fact that we are with a healthy business model and that we are evolving according to our plan, gives them some confidence that organically we are generating capital. You asked about 3 points. Models-DTAs and pension funds. Pension funds, we have to watch it closely. Our information is as good as yours in terms of the evolution of the discount rates.
We disclose in our annual report and in our half-year report what is the sensitivity of our capital to the evolution of the discount rate. These discount rates have to be seen as long-term, good quality corporate bonds for the same maturity of the pension fund. Our pension fund has around the 17-year maturity. Typically, auditors look at A+ or AA- maturity. One has to feel comfortable whether the 1.6% discount rate is the adequate discount rate for a bond that is A+ or AA- and has a 16-year maturity. You are professionals in the market, so you may form a view on this. What I can tell you, but you can confirm it, is that just in terms of the evolution, this rate came down from June to August and is recovering since August. We will see exactly where it stands at the event.
In any case, I'm not expecting anything dramatic there. In terms of the models, this is always a dynamic discussion with the regulators. We have some positives and some negatives. Typically, the regulators always push for add-ons and for more conservative models. We view our models always as probably more conservative than what we see elsewhere. Of course, we don't have as much information. What we know is that we have a very high risk-weighted asset density on one hand, and that we also base our PDs and our LGDs and our statistical data on the recent history of Portugal. As Portugal also improves and as the economic cycle also improves, there is also a benefit to these models. We expect over time the positives and negatives to broadly compensate each other.
We don't know exactly when one will compensate the other, but over time, we expect that the excess conservativeness that our models have in terms of add-ons, in terms of adjustment to the cycle, in terms of the contamination by the recent economic crisis in Portugal that is not contaminating the future data anymore, we expect this to broadly compensate the negatives. DTAs. In DTAs, we have here, basically to make it simple, two types of DTAs. DTAs that are discounted from capital and DTAs that are guaranteed, so to say. The DTAs that are guaranteed, they are real capital, so to say, and effectively what they say is that it tantamount to a state guarantee on this balance sheet item, to make it simple. These are the ones that if they come down, they have a negative impact on the capital ratio. We don't expect them to come down.
Okay. Then there are the DTAs that are not that guaranteed, like the tax loss carry-forwards. The tax loss carry-forwards, they are already deducted from capital. Even if there is, and this is not something that I would anticipate, but just in theoretical terms, even if there is a write-off of non-state guaranteed DTAs, considering the limits that we already have, they would not be deducted from the capital ratio. They would be irrelevant from a capital ratios standpoint. Okay. I believe I've answered your question.
Yes, definitely. Thank you.
We have another question that comes from the line of Maksym Mishyn from JB Capital Markets. Please ask your question.
Good afternoon. Here's Max from JB Capital. Thanks for the presentation and for the Q&A. I have three questions, if I may. The first one is on costs. How much of one-off costs do you expect for the rest of 2018? If you could break them down to Poland and Portugal, that would be fantastic. The second one is on legal provisions. Could you kindly provide us with an update on the fine you received from the competition authority in Portugal and how you plan to provision for it? Finally, the third one is on performance of corporate restructuring funds. The industry segment seems to be doing well, yet the real estate tourism assets remain flat for a while now. Could you remind me of what is inside these assets, and why is it proving more difficult to reduce them? Thank you.
Okay. Thank you very much for your questions. In terms of costs, in terms of the full year cost impact, just from an annual perspective, in terms of recurrent costs, adjusted for Eurobank integration, I would expect them. We have given exactly what is the non-recurrent part of the cost. You also have what are the recurrent costs. I would expect this pace of recurrent costs to broadly continue until the year ends. In terms of the competition authority, we are quite confident that we are right. We have a fine, as you know, of EUR 60 million. We have discussed this fine. Basically just for you to know what's at stake is the following.
Because there was not enough statistical data in Portugal, the Portuguese banks, for some time, they have exchanged statistical information in terms of market shares and in terms of statistical data in terms of market shares on one hand, and they have also exchanged some data in terms of what are the public prices that they would, the pricing tables that they would use mainly for mortgages. This second one was already decided. It should not be seen as something to be challenged by the competition authority. In terms of the first one, to exchange information in terms of market share and to do this statistical data is not something that somehow frames the price or tantamount to price collusion.
Just to give you an information, the Portuguese SEC, so the CMVM, gives every month the market shares of the different brokers, just to give you an information. If this will attempt amount to helping on price collusion, certainly it would not make it. The Portuguese, I don't know, energy authority gives every month market shares of what are the energy companies, what are the market share of the energy companies. We don't see this as, we think that the competition authority clearly exacerbated its position. In any case, as you know, based on the IAS, what the IAS say is according to IAS 37 is you should only provide for contingencies when you see that it is more likely than not that you will pay them. Okay?
Contrary to what happens in IFRS 9, where you provide based on an expected loss, that if you have a 10% risk of losing something, you should provide the value multiplied by 10%. In terms of what are legal contingencies, there is not this possibility. Even if I thought that I had a 5% chance of losing them, I could not provide 5% of it because this is not what the accounting rules say. We are not expecting to provide this based on information that we have because we think we are right. Okay. In terms of the funds. The funds are managed by third parties that try to maximize the value. It is true that an important part of the fund is tourism and real estate.
Effectively, the managers of these funds, they have important equity kickers, and they have their incentives on two sides, so to say. They have on one hand, an incentive on having more capital gains because it goes, as it would expect, directly to their incentive fee. They also have an incentive to sell the assets quickly. The trade-off, if you want, between the speed at which they sell the asset, for which they have a special incentive, and the capital gain that they make is something that's managed by them. We are not managing the funds because we could not. If we were managing the funds, we could not take the assets off balance sheet. It is true that tourism is performing well. We have around 54% of the funds in terms of fair value in tourism assets.
We have around 32% in terms of real estate. Summing both together, we have here around 86% of the funds in tourism and real estate. We know that the managers are managing the funds and have put some assets for sale at prices that they think that they are fair. We would expect during next year to have an important reduction in terms of the real estate funds, but w e are not managing these funds. We would expect to start seeing, exactly because the prices are getting to the levels that make, so to say, these managers satisfied with their return, to have an important use on this regard. We are not managing the funds. Okay. The good news is exactly as you say, is that 86% are in parts of the Portuguese economy that are performing relatively well right now. Okay.
Thank you very much.
Your next question comes from the line of Carlos Peixoto from the CaixaBank BPI. Please ask your question.
Hello. Good afternoon. Carlos Peixoto from CaixaBank here. Three questions on my side. First one would actually be related with Angola in the kwanza devaluation that we've seen in what we had so far of Q1. I was wondering whether that will translate into any additional one-off impacts in the fourth quarter, and what type of amounts could we be talking about? Second question on the evolution of NII in Mozambique. On a quarter-on-quarter basis, we had a significant improvement in NII in Mozambique. I was wondering if you could shed some light on the drivers behind that improvement. Finally, as a third question, on the NPE reduction. From the pace that we are seeing so far, I guess that the 25% target reduction of NPEs in Portugal for year-end or for the full year is going to be well beaten. How do you see it?
Where do you set the target now for year-end? Basically, should we expect that BCP could reach a net NPE ratio of around 2.5% during 2020? Here, I'm basically alluding to the comments on the banking union made by the German finance minister, just to see how do you see asset quality evolving over coming, and attached with that cost of risk.
Okay. Thank you very much. As you know, in terms of Angola, the currency of Angola has depreciated, but now it has recovered the largest part of what it has depreciated. Basically, right now, we are not expecting any important effects related to this. In any case, I can tell you that we don't consolidate the Angolan unit, the Angolan participation enters into the same thresholds as all the participation that we have. It is already, so to say, deducted for capital ratio purposes. Any evolution in terms of kwanza does not affect our capital ratio, okay? There has been not an evolution, even if there were an evolution, it would not affect our capital ratio because the participation is already above the threshold. As you know, we also have the insurance company and so on. In terms of the Mozambican NII.
In Q3, as you know, as you comment, there has been a very good performance. A part of this performance, the Mozambique unit is not so large, but a part of this performance has to do with recovery of interest, so to say. I would not see the Q3 performance of the margin Mozambique to be the recurrent one. I would rather see the Q1 and Q2 more as the recurrent basis. Of course, there will be quarters in which we will recover past due interest that may then have a positive impact in terms of our NII, but there's an element there that is not totally recurrent, at least on a quarter-by-quarter basis. If you do it over the year, we can have few positive news, but that's it. In terms of the reduction of NPEs. We do not constantly update our plan.
As I've said in page 52, we have presented a plan where we would reach EUR 3 billion by 2021. It is true that we are getting there at a faster pace than what we were originally envisaging. We have been doing so without an additional cost. I would not like, and I am not, so to say, mandated to review the plan, but I think it is to be expected that we reach this level somewhat sooner. Exactly when? We have here, [Raul], an issue of much more granularity of our NPEs, because as you know, in terms of the NPEs of mortgages and of private individuals, they are already quite low, and our NPE ratio is already comfortably below the 5% that you mentioned. Basically, we have much more bulky cases.
By having more bulky cases, as I have commented, we will have more volatility in terms of the performance in any given quarter. In any given year, we expect to get the 35%, but we will have quarters where we will have probably less quality in terms of recovery, more write-offs, as we had, for instance, in Q2 of this year. There will be quarters in which we really receive cash or receive assets that we then sell, which as it happened in this quarter. Because the assets are effectively more bulky. I do expect that in any given year, we will reach these targets. The 25% per year seems to be a reasonable target. We are at 37% right now. Going forward, it seems to be a reasonable target.
We will do our best to come to a low level of NPE without destroying several, or a lower value of NPE, without destroying too much shareholder value to get there, as you are seeing. I would not like to anticipate a review of the target.
Thank you. Your next question comes from the line of Gabor Kemeny from Autonomous. Please ask your question.
Hello. My first question is a follow-up on the Polish FX mortgage situation. Some of your Polish peers mentioned that they expect to create a provision for the possible legal cases at the Q4 stage. Is this the approach you expect to pursue? What do you think are the drivers for determining the size of this provision? The other question is on Portuguese NII, where we saw some encouraging trends in the first quarter. How would you expect your Portuguese NII to evolve from here? In particular, if you could comment on two details, if you could give us an update on your interest rate sensitivity to Euribor . An update on how much NII is left coming from the unlikely to pay loans.
Okay. Thank you very much. In terms of the FX mortgage, what I can tell you is the following. We will strictly abide to the IAS. We don't invent the accounting rules. What we will do is the following. We will, at any given time, based on objective information, based on objective data, try to understand whether it is more likely than not that we will win or that we will lose. If and only if it becomes more likely than not that we will lose something, we will have to provide. These are the rules. As I told this a couple of minutes ago, we have no objective data that leads us to this conclusion. As I told you, we have won all the cases until now. At least we have not lost one single case.
There is one single case where the customer has won in second instance, but w e are still challenging it on the Supreme. All the other cases we have won. The information that we have about the competitors is that they have won 90% of the cases. Even after the 3rd of October, we don't have any consensus on how the courts are deciding. We don't have any objective information that would lead us to provide anything. In any case, what we will do at any given moment is to apply the correct accounting rules. That's what I can tell you.
Sorry, does this mean that you will apply a case-by-case approach and not a portfolio-based approach for the FX mortgage provisionings?
We will apply the correct accounting rules at any given moment. That's what I can tell you. Right now, we don't have any objective data that would lead us to make any type of provision. Okay. If the correct approach is a case-by-case. We will apply the correct approach. We don't invent the accounting rules, and the accounting rules are not discretionary. Based on the information that we have right now for our cases, we don't see the need to provide anything. However, these are the 30th of September accounts. This opinion came on the 3rd of October, and it's very early days. Based on what we see right now, we don't need to provide. It's early days. Let's see once we have more information. Okay.
Okay.
We are correct, we are transparent, transparency for us is above everything else. In any case, this is a point that I would like to highlight. We are the controlling shareholder of Bank Millennium in Poland. Bank Millennium is listed, has its own board, is listed there as its own company, this will be, in any case, a decision of the board of Bank Millennium. I think it's important also to say this. Okay, in terms of the Portuguese NII and the Portuguese NII sensitivity, just to start with this. We typically have an NII sensitivity of, for every 100 basis points of reduction of interest rate. Typically, we have, I would say, between EUR 50 million and EUR 150 million of net interest income sensitivity. We manage this value based on our view on where the interest rates are going.
Right now, our institutional view in terms of interest rates, and our view here is as good as yours, to be totally humble and honest, is that we will have a lower for longer interest rate scenario. We are closer to the EUR 50 million than to the EUR 150 million. On average, we are around the EUR 100, as I have here commented several times. Okay, in terms of the Portuguese NII. What we are trying to do in terms of the Portuguese NII is to protect it as much as possible. In this lower for longer scenario, it is difficult to maintain a stable NII or a slightly growing NII in this scenario. What are we trying to do to address it? We are continuing our reduction of term deposits rate, ensuring the convergence between the back book and the front book.
There is still some room to go there, not much, but there is still some room to go there. We are developing more our personal loan business, that is a higher margin and fixed rate segment. We are expecting this to counterbalance these lower for longer trends. Of course, this also depends on the competitive effects and so on, and on the exact, because we are still speaking sometimes of basis points, on the exact basis point at which the Euribor is. We are trying to defend our NII so that it is stable or grows at a high single-digit rate. Okay.
Understood. Thank you.
Okay. Thank you.
There are no further questions at this time. Please continue.
Thank you very much for your interest and your time in following our equity story, and we really appreciate the time you dedicate to analyze us. We are totally committed in terms of transparency to the market and relationship with analysts and with investors. We are here confident that in this next quarter, we will also deliver on what we need to achieve our plan. Let's see each other together in the next presentation. Thank you very much, ladies and gentlemen.
That does conclude our conference for today. Thank you for participating. You may all disconnect.