Good morning to everyone, and welcome to Unicaja Banco third quarter 2018 results presentation. As we usually do, let me start confirming that we have published the quarterly financial report and this presentation this morning before market opens in both the CNMV and our corporate website. Today, we have our Chief Financial Officer, Pablo González , that will explain the results and the main trends of the quarter. After the presentation, we will answer the questions received through the webcast and in the IR inbox. I leave the floor to Pablo.
Thank you, Jaime. Good morning, and thank you all for attending the presentation. I will start on page four with a summary of the results. First of all, let me say that these quarter results are very similar to what we saw in previous quarter, although with some seasonality typical of every third quarter results owing to the summer effect. Regarding the business, lending trends were pretty stable. As you all know, every second quarter, we have close to EUR 400 million of seasonal loans. Excluding such seasonal advances, private sector performing loan book decreased 0.3% quarter-on-quarter. New loan production continues to increase in SMEs and individual segments. In corporate and public sector segment, the new production did not grow due to our proven pricing policy.
Total customer funds continue to grow close to 2% year-on-year, especially inside accounts where balances grew 11% and off-balance sheet funds growing close to 5%. Regarding results, net interest income grew close to 5% in the nine months of the year. Fee income, adjusted by the full acquisition of the insurance company, Unión del Duero, also grew by almost 3%, while total cost fell by 3%. Loan loss charges and foreclosed assets provision trends remain well below the previous year. On asset quality, liquidity, and solvency, our NPA fell by 15% year-to-date or above 20% year-on-year, representing a decrease of almost EUR 1.1 billion. On liquidity, as you all know, we continue to have a very comfortable position with the loan-to-deposit ratio at 74%.
Regarding our solvency position, the CET1 phase-in was 15.6%, and in fully loaded terms, reached 13.7%, or 13.8% when considering third quarter retained earnings, showing a significant improvement during the last quarters. It is important to highlight the significant buffer of capital above our SREP requirements. We have an excess of EUR 1.7 billion in CET1 and close to EUR 1 billion in total capital. On top of this, let me remind you that in September, we have also legally merged with EspañaDuero and finished successfully the IT integration. I will move now to the results and business section with the P&L details on slide six. Net interest income fell 1% quarter-on-quarter in the third quarter. Net interest income in the first nine months of 2018 was 5% above the same period of last year.
Fee income was stable both in the quarter and in the first nine months of the year, although growing almost 3% when adjusted by the full acquisition of Unión del Duero. Regarding dividends, as we said last quarters, there is some seasonality with a big bulk of the income concentrated in the second quarter of the year. The P&L contribution from associates was slightly above second quarter. It is worth noting that this line is not including the contribution of Unión del Duero since first quarter this year. Trading income in the third quarter 2018 was also slightly above the previous quarter, but not as strong as last year. Bear in mind that trading income should increase in the fourth quarter 2018 in order to compensate the restructuring costs. Other operating income and expenses this quarter include EUR 5 million from our real estate transactions.
Regarding costs, they were pretty stable in the quarter and decreasing by almost 3% in the first nine months of 2018 compared to the nine months of 2017. Total impairments reached EUR 36 million this quarter, confirming very positive trends, something that enabled us to reinforce other provisions one more quarter. Total provisions in the first nine months of 2018 were significantly below the previous year. All this left the profit before taxes at EUR 52 million in the third quarter, reaching EUR 192 million in the first nine months of the year, which is almost 50% above the previous year. Attributable net income also grew 5%, from EUR 136 million in 2017 to current EUR 142 million. Now, if we move to customer funds in slide seven, you can see that total customer funds grew close to 2% year-on-year, with off-balance sheet funds up 5%.
Regarding on-balance sheet funds, the mix between sight and term deposit was more stable this quarter, with sight deposit representing 77% of total private sector deposits. In slide nine, we show the credit and loans trends. Total gross loans fell 3% year-to-date, with the big bulk of the decrease explained by lower non-performing loans and public sector loans, decreasing by 18% and 16%, respectively, in the year. However, private sector loans were pretty flattish in the same period. By segment, as you can see in the bottom left, mortgages continued to fall by 3% year-to-date, explaining most of the private sector loans decrease in 2018. In the right side of the slide, we show the performing loans evolution. While public sector loans fell 16%, private sector performing loans was pretty stable year-to-date.
By segment, in the bottom right, you can see that mortgages continued to decrease by 2%, but the quarterly drop is improving every quarter. In large corporates, the portfolio grew more than 4% year-to-date, and in consumer and others, the trend was more stable. Although if we consider only pure consumer loans, balances grew 9% year-to-date. All in all, despite the third quarter relative weaker seasonality, performing loans trend continued to be positive. As you can see in the bubbles on the top of the chart, the year-on-year decrease is lower every quarter. One year ago, the balances were decreasing about 3%, and in this quarter, the drop was 1.5%. In slide nine, we show the new loan production trends. New production in individuals and SMEs continued to grow by 24%.
In public sector and corporate loans, the new production was below the previous year, among others, owing to the significant new production in corporates formalized in 2017, but also due to the prudent and conservative commercial strategy with focus in pricing. Something that is also reflected in loan yields. As you can see in the bottom of the slide, the yield of new loans grew by 10 basis points in individuals and by 12 basis points in SMEs and 13 points in corporates. In slide number 10, we start to review the P&L. As you can see, net interest income fell 1% quarter-on-quarter, while net interest margin decreased only one basis point. The big bulk of the decrease, as you can see in the top right, was explained by the lower income from mortgage with floors. Although such an impact was below the previous quarters.
Bear in mind that one year ago, this impact was close to EUR 4 million each quarter, while in this quarter it was EUR 1.4 million. Regarding the customer spread, in the bottom of the slide, you can see that front book spread remains well above the back book, among others, owing to the pricing policy followed by the bank. In slide 11, we have the regular info regarding the debt portfolio. The size, excluding forward sales, was similar to the end of last year, and the big bulk of the portfolio remained sovereign bonds classified in amortized cost portfolio. Regarding its contribution up to net interest income, it was pretty stable this quarter at EUR 59 million. If we move to slide 12, you can see fee income trends that were pretty stable this quarter. Something positive because second quarter 2018, we had close to EUR 3 million of seasonal fees.
On top of this, following the full acquisition of Unión del Duero in first quarter 2018, there are some fees that are not considered in this P&L line anymore, implying that this quarter fees were positive. In the first nine months of 2018, fees were almost flat compared to the first nine months of 2017. When we adjust the Unión del Duero impact, fees grew almost 3% year-on-year. Moving to cost, as you can see in slide 13, operating expenses fell 2.5% in the first nine months of 2018 compared with 2017, and 6% compared with 2016, driven mainly by lower personal expenses. As we said in previous occasions, we expect to continue to reduce personal costs in the future. Although these will be partially compensated by higher general expenses. As you all know, we're trying to reach an agreement with the unions to further reduce personnel costs.
We cannot elaborate further on this matter, but we expect to have more details next quarter. As we said in the past, our idea is to mitigate the restructuring costs with the realization of gains from our debt portfolio. In slide 14, we have included impairment details. As you can see, total impairments improved significantly compared to 2017. In the first nine months of 2017, we booked more than EUR 200 million of provisions, including the ones related to the reorganization of the life insurance business of the group. This has fallen to just EUR 77 million in this year. It is worth noting that in the previous quarters, we released some loan loss charges, but this was partially explained by the disposal of small written-off portfolios that enabled us to release EUR 18 million in the first half of the year. We haven't sold written-off portfolios this quarter.
As you can see in the right-hand side of the slide, the annualized cost of risk year-to-date, excluding such disposals, was just six basis points, well below the previous years. As we have said in the past, such a positive trend is a result, among others, to the relative high coverage level of the bank and the continued market improvement. If we move now to the asset quality in slide 16, we have details on the evolution of our non-performing loans, where you can see that the trend remained positive. Non-performing loans continued to decrease one more quarter, representing almost a EUR 500 drop year-to-date. In the bottom of the slide, you have quarterly NPL variation, where you can see that the pace of reduction has accelerated further this quarter.
Gross entries fell significantly in the third quarter this year, something that mitigated the lower recoveries owing, among others, to seasonal factors, explaining the 5% quarterly drop in non-performing loans balances. As we usually do, in slide 18, we update our non-performing loan coverage position that remained pretty stable at 55% one more quarter. As we show in the right of the slide, and as you all know, 92% of our non-performing loan balances are secured, meaning that this 55% coverage is a very conservative level. If we move now to slide 18, you will see an update on the foreclosed assets trends. In the left side of the slide, we have the coverage details. Overall coverage remains stable at 64%. On the right, you have the provision released and outflows details.
Year-to-date, we have released EUR 55 million of provisions from disposals, representing 32% of the net book value of the asset sold. In the bottom, we saw that outflows in the third quarter reached EUR 182 million. All in all, these are positive trends as a result, as I said, of our coverage levels. In slide 19, we summarize the overall problematic exposure. As you can see, we have reduced by 15%, or by close to EUR 700 million year-to-date, the non-performing assets gross balances to EUR 3.9 billion. NPA coverage remains stable at 59%. In net terms, foreclosed and non-performing loans balances represent only 3% of total assets. In the bottom of the slide, you can see that the Texas ratio continues to decrease every quarter, reaching now 63%. In terms of liquidity, you have the details in slide 20, where you can see very little changes.
Our loan-to-deposit ratio remains one of the lowest of the sector at 74%. On the top right, you have the liquidity ratios that also continue to be among the highest of the sector. In terms of wholesale maturities, let me just remind you that the most significant impact will start in 2019 and the beginning of 2020, when we have relative expensive maturities that will be very positive for net interest income trends. Finally, regarding the amount of liquid asset, as you can see in the bottom left, they continue to represent more than 25% of our total assets. All in all, this shows a very comfortable liquidity position for the bank. Finally, moving to solvency. In slide 21, you have the details. First of all, bear in mind that we do not include in the reported capital ratios for Q3 2018, the retained earnings.
As you know, we only include retained earnings in the second and fourth quarter of the year. As you can see in the top left, current regulatory capital ratios are well above the SREP requirements, with a EUR 1.7 billion buffer over CET1 and close to EUR 1 billion buffer in total capital. CET1 phasing reached 15.6% in September, while total capital was 15.8%, both of them 30 basis points above the previous quarter. As we have explained in the previous occasion, for us, the regulatory solvency is very relevant because the phasing calendar that we apply finishes in 2023, rather than the most common 2019 deadline. In other words, our regulatory capital ratios will not meet the fully loaded ones until 2023. If we look at the fully loaded terms, the CET1 reached 13.7% or 13.8% when considering the third quarter retained earnings.
This ratio was 13.4% if we exclude the unrealized capital gains from our debt portfolio. As a reminder, take into account that gains will be released to compensate restructuring costs probably next quarter. We have included our credit risk weights in the bottom left of the slide. As you all know, we continue to apply a standard approach and as a consequence, very conservative credit risk weights. Something that further reinforces our solvency position. Let me finish, as we usually do, the presentation with some final comments that you have on page 23. As I said, we continue to report resilient results generation capacity one more quarter with the bottom line 5% above the previous year. The commercial activity continues to improve following a prudent pricing approach.
It is also worth noting that the non-performing assets continue to decrease at a very positive pace, among others, owing to a best-in-class coverage of the problematic exposure. Finally, as we have been doing since we become listed, we are generating capital and keeping an extraordinarily comfortable liquidity position quarter after quarter. Thank you very much, and I will now answer your questions.
Thank you, Pablo. Moving to the questions, we will start to probably expect one on the mortgage stamp duty tax. If we can provide the expected impact, our views on the situation, what is the expected outcome from next Monday? What is our commercial strategy? If you can elaborate a little bit on the matter, Pablo, please.
Okay. All that we can say is that this is a tax issue. It is not a mis-selling process. We have been following the fiscal law during the last 23 years as the entire sector has done. We have followed the law and also the interpretation made by different courts. Under our view, retroactivity doesn't apply in this case. The law stated always and in a clear way that the client was the one paying this tax, like in other European countries like France, Italy, Luxembourg, Portugal, or Austria. In the case of Spain, that is why there are exemptions depending on the age of the client, number of family members, and other, something that can only apply if the client is the one that has to pay the taxes.
It is also worth noting that subrogation law specifies that this tax is exempt, and obviously is exempt for the client. Other examples is that this tax is deductible when calculating the capital gains following the disposal of the house. This is an issue related to the tax that was paid by the client to the regional governments following the law. The Spanish law also says that the application of this type of ruling should be taken into account since its publication, so without retroactivity. This is the meaning of the principle of legal certainty and the fundament of our rule of law. We really need legal certainty and clear rules to follow. Something that is very important in mortgage contracts, which are long-term contracts. We expect to have a final confirmation in the coming days, and hopefully will reduce the current uncertainty.
Until then, we prefer not to anticipate any internal decision or impact. We are monitoring closely the situation, and we will adapt our strategy to whatever outcome we have from this matter. We will continue to price mortgages considering the competition and the internal costs, as we always do. So far, you can imagine, we cannot be more specific.
Thank you, Pablo. The next one is on mortgage floors, if we can update the situation regarding the mortgage floors.
The balance of mortgages with floors continues to decrease. Balances with active floors were above EUR 3 billion one year ago, and now they're close to EUR 2 billion. In the quarter, lower balances with mortgage floor had a negative impact, as I mentioned, in NII of around EUR 1 million, well below the impact we used to have in the past. In terms of provision, we have around EUR 300 million for all these legal matters, with around two-thirds related to the mortgage floors.
Thank you, Pablo. The next one is on the IRPH, if we can provide how many mortgages we have linked to the IRPH index and also other legal issues or other legal matters that concern us. Thank you.
Regarding the IRPH reference, this is not really something material in our case. We have around only EUR 200 million. We don't expect any relevant impact from this. Regarding other risks, as you saw in recent quarters, we have reinforced this type of provisions following a prudent approach. We have booked close to EUR 80 million of other provisions year to date, mainly related to all these issues. As I said before, we currently have close to EUR 300 million of provision for legal matters.
Thank you, Pablo. Moving to the ALCO portfolio, if we can update or elaborate a little bit on the size, the strategy, the contribution, and the exposure. The sovereign exposure, sorry, Pablo.
Okay. The size of our debt portfolio, net of the forward sales, has been quite stable year to date. We did some opportunistic acquisitions throughout the quarter, but nothing very significant. As we explained in previous occasions, the amortized cost portfolio size and duration is managed depending on the structural liquidity position of the institution, taking also into account the market conditions. For the fair value to OCI, this portfolio, we manage them actively, so the size and risk sensitivity of the portfolio will depend on our view of the evolution of markets. In terms of exposure, the sovereign debt represents around 78% of the total debt portfolio, which is most being Spanish debt.
We also have some BTPs of around EUR 100 million in the fair value OCI and around EUR 3.6 billion in the amortized cost portfolio with an average duration of less than three year, and no impact in capital or P&L. We also have EUR 750 million of Portuguese debt in the amortized cost portfolio, also with no impact, obviously, on capital or P&L. The remaining sovereign exposure is all Spanish sovereign debt. Regarding the contribution to NII, as we explained in the previous quarters, as soon as the forward sales are executed, probably in the next quarter, the contribution of the portfolio will get back to levels similar to those in last year, where quarterly contribution was around EUR 55 million per quarter, which should be a more normalized contribution.
Thank you, Pablo. We also got a question on the interest rate sensitivity. We can update the sensitivity on interest rates.
We remain positively positioned towards an increase in interest rates, especially for the bucket for the 2020, 2021. The impact in the 12-24 months bucket of NII from a parallel increase of 50 basis points on the rate curve as of today is slightly higher than the 6%, compared with the NII expected in that period of time with the current implicit rates. If we compare the NII for this bucket, the 12-24 months bucket in two years' time, calculated with the current implicit rate plus a parallel shift of 50 basis point, the impact over the actual NII rises up to 12%. As we mentioned in previous quarter, this calculation is considering what we think is a prudent and realistic assumptions like, for example, the changes in the mix between term and sight deposit as rates go up, and a constant balance sheet.
Thank you, Pablo. Another one related to NII, if we can update the guidance for NII for this year and next one.
As we said in previous occasions, we expect the NII to grow this year. Once we realize the gains from our fair value to OCI portfolio, the interest income from such portfolio will be lower. Despite such an impact, we continue to expect NII of 2018 to be above last year. For next year, we also expect to have the NII above this year number.
The next one is regarding the TLTRO. We got several questions on the date of maturity and an update on the situation, the funding position with TLTRO.
All our ECB funding that we currently have is related to the TLTRO II facility due to our strong liquidity position and net interbank liquid position. We decided to go in full to the ECB TLTRO II facility in full in the last auction. The maturity of this will be in March 2021. In terms of liquidity, we won't have any stress because the amount that we'll redeem from this TLTRO has been invested in sovereign bonds with the same average maturity and in a very similar amount. On top of that, our NSFR ratio and LCR ratio are high enough to offset the deduction of the one year previous to the maturity of the TLTRO. We don't expect any impact in terms of liquidity.
In terms of NII, the positive fact here is that this maturity of the ECB facility will come after the maturity of EUR 1 billion of cover bonds at an average cost of 2.5%, and the maturity of what we call the peaks that are EUR 1.3 billion of high yield deposits at 4.3% cost. By 2021, we expect to reduce significantly the cost of funding, well above the reduction on interest income that will come from the maturity of the fixed income portfolio matched with this TLTRO funding.
Okay. Thank you, Pablo. Moving now to volumes, if we can elaborate a little bit further on the expected loan growth. Thank you, Pablo.
As we have been saying lately, we expect to see the inflection point in the coming quarters in terms of loan growth. As we explained in the presentation, performing loans decrease is lower every quarter, and excluding the seasonal advances, the drop in this quarter was just 0.3%. The third quarter is usually a weak quarter owing to the summer break, so the trend remains positive, especially in individual segments that is improving every quarter. As you all know, we need to see some stabilization from the mortgage book in order to see the overall loan book growing. The new mortgage production is growing every quarter. In this third quarter, performing mortgages fell only EUR 69 million compared with the previous quarter, which represents 0.4% quarter-on-quarter drop and compares with 1.4% decrease in the same period last year.
The trend continues to improve every quarter, and we reiterate our guidance for an inflection by the end of this year or the beginning of next year, when we expect balances to start growing at least at low single-digit rates.
Thank you, Pablo. The next one, it's on the cost-cutting plan, if we can provide more color regarding the negotiation with unions.
As you probably know, we are currently discussing with the unions different options and measures. As you can imagine, we cannot elaborate at this time more on this matter. What we can confirm is that we want to further improve our efficiency, we expect to have more details in the coming months. The idea is to update and provide more details in 2018 annual results presentation. Until then, we cannot elaborate further on this topic.
Thank you, Pablo. The next one is on impairments, cost of risk, and future impairments going forward.
Let me start with the loan loss charges. In this first half of the year, we sold some small written-off portfolios, this, as I said, explain around EUR 18 million of provision released. We haven't sold this type of assets in this quarter, as I said. Excluding such disposals, the cost of risk was only six basis point. This, obviously, is a consequence of our strong coverage levels. Regarding foreclosed assets, as we anticipated in the past, following the update in the price of last year, we don't expect these provisions to have any significant this year. Another good news probably coming from both the loan loss provision and foreclosed assets. Finally, on other provisions, we were below the previous quarter, but continue to include some legal impairments.
In the fourth quarter, these other provisions will be much higher, because we expect to book the restructuring charges related to the cost-cutting plan that we mentioned.
Thank you, Pablo. Moving on asset quality, we can update the non-performing asset balances targets and potential portfolio disposals.
We continue to have the same strategy, which is to continue reducing balances while preserving our shareholders' value. We will continue to analyze any alternative, including portfolio disposals, as we have done so far. However, it is worth noting that this strategy has some volatility in the trends and speed of reduction. For example, in the second quarter this year, the big bulk of the decrease came from the non-performing loans. This quarter, the decrease was much more balanced between foreclosed assets and non-performing loans. The pace of decrease won't be linear, but we expect to continue reducing the balances. Bear in mind that in the time of our IPO, in the second quarter 2017, we published a target of reducing from EUR 5.4 billion to EUR 3.5 billion of gross non-performing assets by 2020.
Just five quarters after that, we have already reduced them by EUR 1.5 billion to the current EUR 3.9 billion. The trend and the results from such trend have been very positive, and we continue to be having good news in this front in the coming quarters.
Moving to solvency. There's one specific question on if we can explain why our solvency phasing calendar finishes in 2023.
Yes. I think this is important from the solvency point of view. For most of the banks, the phasing calendar will finish next year. Next year, the fully loaded ratios will become the same as the regulatory ratios for most of the banks, but not in our case. For Unicaja Banco, the solvency phasing calendar finishes in 2023. This is because we apply for the long calendar, something that is explained owing to the acquisition of EspañaDuero, a bank that had a formal restructuring plan approved by the European authorities. For banks with this type of restructuring plan, the phasing calendar applied is the loan one. We will continue applying a phasing calendar until 2023. For us, the phasing is even more relevant than for anyone else.
Thank you, Pablo. On the Core Equity Tier 1 fully loaded, they're asking us that it continues to grow, if we can provide an update on the dividend policy.
As you probably know, we have a CET1 fully loaded target of 12%, while our current ratio, excluding the debt portfolio gains, is 13.3. One of the highest of the sector and calculated with the standard models. We have a very comfortable solvency position. Regarding dividends, I cannot confirm now this year final payout, but we expect to do so in the next quarter in the 2018 annual result presentation. Obviously, as you know, we have a lot of room there.
Thank you, Pablo. One final, actually two final questions that I put them together. One, because they are related. One is expected on the TRIM impact, if we expect some kind of impact from TRIM, that's straightforward. The other one is on the situation of the approval of the IRB model, they're linked.
Regarding the TRIM, we don't expect any impact, because we continue to apply the standard model, so we don't have any impact from this process. On the IRB models, we continue to work in the project. As we stated several times, this is a medium term issue, and we still don't consider this in our capital management at this moment. In the medium term, obviously, either the IRB or Basel IV should be quite positive news for us, but we'll need to wait till they become a reality. This is not a short-term issue, but obviously, if anything, it will be positive news for us.
Okay. Thank you, Pablo. Thank you all. Thank you very much for attending the webcast. Please do not hesitate to contact the IR team for further questions. See you next quarter.