© 2024 Alpha Spread Limited. All Rights Reserved.
www.alphaspread.com is not operated by a broker, a dealer, or a registered investment adviser. Under no circumstances does any information posted on www.alphaspread.com represent a recommendation to buy or sell a security. The information on this site, and in its related newsletters, is not intended to be, nor does it constitute, investment advice or recommendations. The information on this site is in no way guaranteed for completeness, accuracy or in any other way. In no event shall Alpha Spread Limited be liable to any member, guest or third party for any damages of any kind arising out of the use of any content or other material published or available on www.alphaspread.com, or relating to the use of, or inability to use, www.alphaspread.com or any content, including, without limitation, any investment losses, lost profits, lost opportunity, special, incidental, indirect, consequential or punitive damages. Past performance is a poor indicator of future performance.
You don't have any saved screeners yet
Intrinsic Value is all-important and is the only logical way to evaluate the relative attractiveness of investments and businesses.
Good day, and welcome to the TCS Group Second Quarter and First Half of 2019 IFRS Financial Results Investor Call. Today's speakers will be Oliver Hughes, CEO; Ilya Pisemsky, CFO; Sergei Pirogov, Head of Corporate Finance. Today's conference is being recorded.At this time, I would like to turn the conference over to Oliver Hughes, CEO. Please go ahead, sir.
Thank you. Good afternoon to everybody. Today, I'll give you an update on the performance of the business in the first half of 2019 and some color on the overall operating environment. The first half of the year was very productive. As you may remember, we made half of our growth guidance in the first quarter alone. As a result, we upgraded our growth outlook for 2019 to 60% plus. We raised $300 million of gross proceeds to support the incremental growth opportunity we identified in 2019, and to boost our capital position beyond this year in a much tighter regulatory environment, with what we're now probably seeing as the highest risk weights in the world. Thank you for your support in this capital round.I'll start off with an overview of significant business developments before handing over to Ilya for the details.Building an ecosystem is all about scale and growing the customer base. In the first half of 2019, we brought 1.5 million new unique customers into the Tinkoff ecosystem. 1 million customers were noncredit and transactional customers. 0.9 million customers were credit customers. You probably already noticed that these 2 figures tally to more than 1.5 million, and this highlights an important trend. A growing percentage of our incoming customers take more than 1 product soon after coming to us.Net loan growth continued to pace into the second quarter, adding another 20% quarter-on-quarter. The credit card book, our mature and largest lending business grew at an-above market rate but slower relative to the newer credit business lines. Strong growth, albeit from a much lower base, came from personal loans, car loans and home equity loans.By the end of the second quarter, the share of noncredit card loans in the loan book had reached 36%, amply illustrating the ongoing diversification of our consumer lending portfolio and the movement to secured lending. The economics of our lending business is very strong, and we continue to book very NPV-positive customers in large numbers.Question of asset quality was widely discussed during our recent roadshow. I'll repeat here what we said during these investor meetings. We believe that we're now past the midway point in the credit cycle. The consumer lending market has been growing strongly for over 2 years, and personal leverage has been building up in the system. This means that we'll no longer see the gradual reduction in cost of risk that we saw over the last few years. Cost of risk probably plateaued at the end of 2018, and going forward, we may see a gradual increase in cost of risk across the market as we move into the second phase of the lending cycle.However, I'd like to emphasize that we're not seeing any major warning signals in our credit businesses. The higher cost of risk that you've seen in second quarter results comes, from a large extent, be attributed to IFRS 9 front-loading of provisions and fast-growing portfolios and the base effect, as Ilya will explain in a few minutes.Our transactional and servicing businesses are on track after a seasonally slow first quarter with revenue of RUB 12.1 billion, which is 30% of total revenue for the group.Tinkoff Insurance grew strongly, showing almost 28% quarter-on-quarter increase in premiums earned. The bulk of this growth came from car insurance, both voluntary, CASCO; and mandatory, OSAGO, sold mainly through dealerships and agents, in addition to our existing online channels.Tinkoff Black reached 5.7 million customers by the end of the second quarter. Current accounts remain central to our strategy of growing our share in the mass-affluent segment and realizing the ecosystem effect through growth in the customer base, multiple digital services and cross sell.Tinkoff Business added another 30,000 new accounts, reaching a total 484,000 accounts opened at the end of June. We continue to build out the SME platform and broaden the product offering, recently adding a cloud-based, outsourced accounting service and term deposits for business customers.Tinkoff Investments are going great guns and hit over 590,000 brokerage accounts opened by the end of the second quarter. According to the Moscow Stock Exchange, in the first 7 months of 2019, Tinkoff outpaced Sberbank by the number of accounts opened.Meanwhile, we're actively expanding the range of services for our investment customers. For example, qualified investors can now participate in IPOs. And I'm also pleased to announce that Tinkoff Investments broke even in July, ahead of plan and despite the heavy investments in growth of the customer base. Tinkoff Acquiring is doing well and was recognized among Europe's top online acquirers, ranking third among Russian banks for 2018, according to the Nielsen report.Other results. The group made net income of RUB 8.2 billion for the second quarter and RUB 15.4 billion for the first half. This represents a year-on-year increase of 36% and an ROE for the second quarter of 64%.As a result of the SPO, we raised $294 million net proceeds, and we're currently working ways to put this money to work to bolster our capital positions while we submit the applications to the CBR. The process of registering the capital increase can take some time.I will go over noteworthy events. I'd like to mention that Tinkoff increased its stake in CloudPayments to 90% from the 55% stake acquired in 2017. In June, we launched the world's first voice assistant for financial and lifestyle interaction, through the mobile app called Oleg. Tinkoff Travel digital platform was launched in the Tinkoff mobile app. We took the tough decision to discontinue our mortgage origination platform. While it was a superb service for customers and partners, due to the structure of the mortgage market dominated by state-owned banks, it was not financially viable as a stand-alone business line.Also in June, we announced the launch of Tinkoff Capital, expected by the end of this year. This will manage or initially offer customers Tinkoff's own ETFs and will be a simple and transparent line of products for investors without interval funds or closed-end funds.In July, Tinkoff made its proprietary Tinkoff VoiceKit speech technologies available to market. External customers can now buy speech-to-text and text-to-speech services from us. This cements our position as a technology player in the Russian market. I'm also pleased to announce that Tinkoff has been named the Best Russian Consumer and Corporate Digital Bank in the ninth -- 2019 World's Best Digital Banks in Central and Eastern Europe Awards by Global Finance magazine.And on that, I'll hand over to Ilya, who will go through the financial and operating results in some more detail.
Thank you, Oliver. Hello, everyone. As you can see on Slide 4, in the second quarter of 2019, our assets grew by 10.6% due to strong net credit portfolio growth in the amount of RUB 49 billion. The net credit portfolio increased its share in the composition of assets to 64% of total assets. Net loans increased by 20% during the quarter, and by 46% since year-end, and amounted to RUB 290 billion.Slide 5, you can see that the gross loan portfolio increased by 19% or RUB 53 billion. 44% of this growth came from the credit card part of the loan book with 620,000 new credit cards activated.Another 24% or RUB 13 billion came from personal loans, and 18% or RUB 9 billion from home equity loans and then from point-of-sale loans and auto loans.Loan portfolio growth will slow down in the second half of the year, but we have little doubt that we will reach our target of 60%-plus loan growth for the year.I will reiterate the parameters of issued loans. They are as follows: The average approved limit size for a new credit card is RUB 65,000. The average balance across the existing portfolio is RUB 69,000, and the yield is 33% on average. The average size of a personal loan is RUB 255,000. Duration is 2.5 to 3 years and yield, 25%, 26%. With the average size of point-of-sale loan is RUB 27,000, duration is 12 months and yield of 25%, 26% as well. The average size of a car loan is RUB 520,000. Contractual duration is 5 years and yield, 20%. Average LTV is 94%. And the average size of a home equity loan is RUB 1 million. Contractual duration is expected to be 10 years and yield 17%, 18% with average LTV of 35%.The NPL ratio continue to go down to 7.4% of the 1st of July. Loan loss provision coverage improved 273% of nonperforming loans, and this, of course, relates to the continuous flat growth of new portfolios but also reflects the stable risk environment.The group's funding structure can be seen on Slide 7. Individual customer accounts increased by 13% quarter-on-quarter, with both current accounts and deposit showing strong growth. SME customer accounts increased by 6% after a seasonal slowdown in the first quarter, as we continue to actively acquire SME customers.We'll also report part of our securities portfolio through the second quarter in the amount of approximately RUB 20 billion. As we attracted roughly the same amount in July during the SPO, we no longer need this additional short-term funding, taking into account that deposits from current accounts money are coming in strongly in July, August. We also reduced our rate for 1-year ruble deposits by 0.5%, in line with the market.On Slide 8, you can see that shareholders' equity increased in the second quarter by 16% to RUB 54.5 billion due to strong growth of the bottom line, the reduction of the dividends outflow from 50% to 30% and the positive regulation of the securities portfolio held available for sale.As a result, our Basel capital adequacy ratio was relatively flat, despite the fast-growing growth of assets. As for the statutory capital adequacy ratios, they were down slightly. N1 went down to 11.1% while N1.1 went down to 7.7%, part from the growth of the loan book, accelerated by high RWA requirements. There were also specific seasonal factors which affected the statutory ratios, annual regulation of operational risks in April and also high annual tax payments for 2018, which reduced statutory quarterly net income.In the third quarter of the year, we expect to see an improvement of capital adequacy, both measured by Basel III and statutory rules. Please note that the SPO was a third quarter event. So in the second half of the year, improvement in the ratios will happen due to the SPO proceeds, the dividend suspension, the pickup in statutory process and some reduction in speed of growth of the loan book.Now some comments on our profit and loss statement beginning on Slide 9. Compared to the first half of 2018, revenue grew by 43% to almost RUB 74 billion, while the share of income from credit products went down in the revenue composition from 71% to 69%. It is worth mentioning that in recent quarters, interest income has been growing a bit faster than noninterest income, mostly due to exceptional growth of the lending part of our business. This does that mean that we are abandoning the idea of further diversification of our business, but this year, credit-related income has been growing extra fast.Interest income for the first half of 2019 grew by 41% year-on-year to RUB 50.3 billion, which corresponds to similar growth in the loan book. Gross interest yield on the credit portfolio went down from 36.1% a year ago to 33.5% due to the change in the loan portfolio composition in favor of less-yielded loans. At the same time, gross yield picked up a bit compared to the abnormally low first quarter, but this should not be viewed as a reverse in the long-term trend. In the upcoming few quarters, we will observe a gradual reduction in gross yield of approximately 0.5% to 1% each quarter, taking us into 28% to 30% area, a year from now.Interest expense for the first half of 2019 grew by 38% year-on-year to RUB 9.7 billion, which corresponded to the growth of the funding base. Cost of borrowing went down from 6.3% a year ago to 5% due to the change in the funding mix towards cheaper current account funding. And the recent pickup in the cost of borrowing reflects the return to the wholesale market with the ruble bonds and a strong inflow of 1-year ruble deposits from individuals. There's also some effect from deferral methodology change, as we allocate some of the salary costs associated with the current account business to interest expense rather than acquisition costs. And this effect gave us approximately RUB 220 million in 2019, as shown in the second quarter.As a result, you can see on the next slide that net interest income for the first half of 2019 grew by 41% year-on-year to RUB 39.7 billion, while net interest margin went down from 24% a year ago to 23% in the second quarter. Our cost of risk remains elevated with 8.2% for the first half of the year compared with 7% a year ago. As Oliver already mentioned, we have passed the mid-cycle point. So in general, we were expecting to see some worsening, but now is the time to discuss this in more detail.There are 2 factors that explain the increase in cost of risk. One of the factors is purely the effect of fast growth and the other is the worsening of asset quality. In order to visualize these factors, let's assume that we grew in 2019 at the same speed as we originally guided for, around 40% annualized. Our planned loan-loss provision grew RUB 8.6 billion in the first half of all 2019, but we provisioned, as you see, RUB 3.1 billion more, including RUB 1.8 billion of extra loan-loss provisions or roughly 1.3% additional cost of risk due to growth higher than 40%. And the remaining RUB 1.3 billion of extra loan-loss provisions or roughly 0.9% additional cost of risk relates to the slight deterioration in portfolio quality.Another way to put this into perspective is to compare how the old portfolio behaves compared to the new portfolio, where new portfolio relates to 2019 issue. Out of 8.2% of total cost of risk for the first half of 2019, 4.4% relates to new issuance including 3.3% from the front-loading effect of IFRS 9 and 1.1% from the deterioration in quality. The old portfolio gives us 3.8% of the loan losses, which is not bad compared with 4% a year ago.What are the reasons for the slight deterioration in quality in the first half? Well, a little bit of everything. Some softening of roll rate, when the debt is harder to collect when a customer becomes delinquent. Some seasonality, as there were abnormally long holidays in May this year. And some operational decisions, well, for example, debt that gets fast approval without verification in some segments of the credit card portfolio.Our fee and commission income on Slide 12 increased by 35% year-on-year and amounted to RUB 16.3 billion for the first half of 2019. All sources of fee and commission income showed growth year-on-year, with acquiring and Tinkoff Investments, which is in other category, being the most rapidly growing segment. In the second quarter of 2019, we have recovered, after the previous quarter where fees were seasonally low. Our current account business brings us more and more customers, which in turn, increases the transaction volumes as you can see on Slide 13. At the end of first half, we had about 5.7 million current account opened with almost RUB 155 billion of balances and almost RUB 400 billion in debit card transaction volume.The continuous growth of customers and balances allowed us to build up fee income, which amounted to RUB 1.9 billion in the second quarter of 2019, net of cash back paid to customers.The current account business is hovering around breakeven, taking into account positive margin contribution from our treasury portfolio into which we deploy cash -- most of the cash that customers hold on their accounts.Our SME business is growing in terms of number of customers, which can be seen on Slide 14. And at the same time, growth in balances and fees improved after seasonal dip in the first quarter but remains lower compared to what we observed a year ago. Due to increased regulatory pressure across the market, we have to adopt a more conservative approach to the business line -- to this business line. There is an expectation though, that in the second half of the year, we will see an improvement in the growth of fee income and transactions.Nevertheless, at the end of the first half, we had over 480,000 accounts with almost RUB 40 billion in balances. We earned RUB 4.2 billion in fees in the first half of the year, in addition to treasury income. On Slide 15, you can see some operational statistics in our service Tinkoff Investments. We see strong customer interest in this product, which plays out in growing number of customers, which was almost 600,000 at the end of the first half of 2019. Fee and commission income grows accordingly despite some introductory trading rates that we give customers as part of our growth effort. Rapid growth assumes high acquisition cost, so Tinkoff Investments was in the red in the first half of the year but turned breakeven in July.Now turning to Slide 16 for some comments on operating expenses, which grew 36.5% year-on-year to RUB 23.1 billion. You can see that the fastest-growing item was the marketing part of acquisition costs, including heavy advertisement on TV and Internet. This part of cost increased 89% year-on-year. The salary component of acquisition cost includes the sale of the diversification of second fastest growing item, approximately 24% year-on-year. Administrative staff costs and other admin costs show more and more of this growth of 21% and 23%, respectively, demonstrating growing economies of scale in most of the business. This was the effect in the cost-to-income ratio, which went down to 41% in the first half of the year and below 40% in the second quarter of 2019 on a total cost basis. Taken without acquisition cost, the improvement in the ratio is stronger to 23.3% or 38 basis points year-on-year.Overall, we were able to show a quarterly profit of RUB 8.2 billion, and almost RUB 15.4 billion for the first half of the year, which is 31% growth year-on-year. Our return on equity is still above 60% despite the growing equity base and fast growth of the loan book, increased loan loss provisions and high acquisition cost.And now back to Oliver.
Thanks, Ilya. So to conclude, I'd like to touch on the regulation first. In June, the CBR announced details on the long-awaited PTI regulation, which will come into force from October 2019. Under this new regulation, unsecured loans with PTI above 50% will be subject to even higher risk weights. Other weight for these loans will be increased by 20 to 100 percentage points. As we previously said, we welcome this regulation. It protects our borrowers from the shelf or lending practices of less rational competitors, and it protects the market as a whole while increasing the barriers to entry into this market.The CBR is also considering increasing risk weights based on the term of the loan. This is also good news because we believe that unsecured loans with a term of longer than 3 years are most likely to turn toxic in the downturn. I'll remind you that Tinkoff does not issue unsecured loans with a term longer than 3 years. While Tinkoff has been using PTI and DTI in its scoring algorithms for many, many years, please be aware that this new regulation will require readjustment from us, as with any new far-reaching regulation.In our approved application inflow, only around 15% are PTI over 50%. As we've explained, there is no direct correlation between risk and PTI. It very much depends on the profile of the applicant. However, we'll take a view as to whether after 1st of October, we decline these customers and, therefore, losing them as a borrower or whether we approve them, but take a hit on capital through increased RWA. This will basically become part of our updated NPV calculations. Please also bear in mind that as the CBL seeks to further cool the market, there's likely to be more regulations.And as usual, I'd like to finish with an update on this year's guidance.We reiterate net income of at least RUB 35 billion. We expect net loan growth to be substantially higher than 60%. Given that our loan growth will be substantially higher than 60%, we, therefore, adjust our guidance for cost of risk to 7% to 8%, mainly due to the front-loading effect of IFRS 9. And the cost of borrowing is reaffirmed at 6% to 7%.Thank you for your attention, and we'll now hand over to questions.
[Operator Instructions] And we can now take our first question from Andrey Pavlov-Rusinov from Goldman Sachs.
It's Andrey from Goldman Sachs. Solid results. I've got, first of all, a couple of questions with regards to your financials and accounting. First of all, if I may, basically on the interest income and the increase in gross loan yield, especially in credit cards. We have seen almost 250 basis points increase in the credit card yield. So I guess this is partly driven by some deferral methodology. Could you explain a little bit more, basically, what has happened, and if it has any far-reaching implications for the second half of the year? Or basically, is it a one-off? Should we completely remove that to increase from the second half base? Or it will still have a supporting effect on the following quarters? And also with regards to the interest expense as well, what exactly changed in your accounting methodology, and whether it again has an impact on the second half?
So for the -- yes, there were some methodological change in the accounting for transactions on credit cards where we basically reduce the -- when we reduce the income via cash back. So we've been basically -- always deferring the income. But cash backs, we always put as the expense, yes, at a time when the cash back is accrued. And right now, we are also deferring some part of the cash backs. So that's the effect. That's what we've done on the interest income in terms of accounting. As I said, we will still have a decline in interest income. And I guess, the effect was quite sharp because of the dip in interest income in the first quarter, where it went down quite significantly after the fourth quarter of last year. What else? The part of this cash backs against interchange, we also have some similar things that -- on deferral of royalty payments.
And so could you...
So total effect -- yes, sorry?
Yes, but basically the total effect Ilya?
Yes. Total effect is slightly over RUB 600 million for the first half of 2019.
But it was all booked in the second quarter, right?
That's right.
Okay. And for the -- basically, on the Slide 10, you also mentioned that there were some salary costs which correlated with the...
Yes. So basically, yes. Methodologically, yes. There were some expenses which we've always been booking in the customer acquisition expense. Decided -- this year, we decided in the second quarter to reclassify -- to basically, to show them instead of customer acquisition costs, to show them in the interest expenses. So again, that's the -- basically, this reclassification happened in the second quarter. So in the first quarter, they were in the customer acquisition expense, but this time, they're in the interest expense. So the total effect is just over RUB 200 million and it's all basically went into second quarter.
Okay. Andrey, do you have any other questions?
Yes, I do, if I may. Basically, also on the cost growth then, given that you made some reclassification to the interest expenses on the customer acquisition costs, but they still grew pretty strongly in the second quarter. Could you explain what was the driver? Because there was one line that looks to have driven the cost growth was the other acquisition costs where there was an increase by RUB 300 million.
Well, in the other acquisition -- yes, in the other acquisition cost, it's basically TV advertisement and also -- well, basically, some acquisition effort. So when we -- for example, buying leads in the Internet, this act of -- this kind of -- mostly that was an increase in TV advertisement.
Okay. And just generally on the cost growth, should we expect the similar growth rates for the second half or there could be some deceleration?
There will be growth, obviously, in the second half, but you should not expect it to grow at the same rate as the first half.
So it will be slower?
Yes, there'll be strong growth in the second half, but we'll be doing less acquisition activity in terms of the credit business lines, there'll be less TV advertising. So you'll see deceleration.
Cool. And just if I may, the last question about SME business. There, there was some -- again, some deceleration in the both the customer acquisition and costs and also in, generally, in the revenues and client balances. Could you give us a sense of what's happening in this business? And what are the prospects?
Sure. So you always get a bit of a seasonal slowdown in the first quarter, in particular, in SME. So if you think where we are, we're not in large, medium-sized corporates. We're in micro business, meeting individuals, persons, the small end of S. So here, you see a seasonal dip. If this continued, basically it was flat still into the second quarter. So this is something -- is more or less market-wise, when you look at all of the players in the segment. And the seasonality was accentuated by, let's say, the ongoing regulatory intervention in the segment across the market. So it slowed down the whole market. We've discussed this many times on previous calls and in meetings before. So this has been going on since basically the late summer or early autumn of last year, and this basically had a knock-on effect. So this has caused us to do a number of different things. We've adopted, as usual, to the various regulatory challenges that come our way as we do, as a large but also very adaptive player in the segment in which we have this existing business. So you saw the signs to recover in the second quarter, and you'll see it continue to recover throughout the year. But also, we've used this as an opportunity to retool and do more in the M of SME, and we're actually seeing quite good traction in terms of attracting new customers. A part of our product build-out is not just to work with the small businesses in the micro segment, but also to attract new customers in the medium-sized business segment. And so things like deposits, for example, to a certain extent, short-term loans of overdraft facilities, to be more precise. These are kind of things that they need, so we're actually booking quite a lot of those customers, and that will drive our growth going forward as well.
We can now take our next question from Andrew Keeley from Sberbank.
I have a couple of questions, maybe take them one by one. Just a bit of a clarification on your -- the comments you made, Ilya, in terms of the cost of risk. That was kind of a flurry of numbers. Am I right in thinking that, basically, in terms of the additional provisions that you guys have been booking this year due to the kind of higher-than-expected loan growth, you're basically saying that around 60% of this kind of additional provisions are due to the kind of IFRS 9 effect, and around about 40% are due to the kind of size of some worsening in asset quality? Is that more or less right?
Yes. Yes. That's right. 60% to 70% is roughly -- is attributed to the IFRS 9 policy and approximately 30% of 40% to the duration of the portfolio.
Okay. Just so a kind of a follow-up on this. When I look at your kind of cost of risk by segments, I think one of the things that's quite striking is there's been quite a big pick-up in the cost of risk on cash loans, really, over last couple quarters. And I think in the second quarter, it was up something like 14%. And if you look at the NPLs on cash loans, they also went up very strongly in the second quarter. I'm just wondering -- I mean, this is obviously a relatively new segment for you guys. Do you have any concerns here? I mean why are we seeing this very strong pick-up in the cost of risk on that product line?
Well, that's basically this -- there's a smaller base in the subsegment of our loan portfolio. So IFRS 9 effect is higher on cash loans compared to credit cards. With -- well, and it's basically -- it is in the picture because the cost of risk -- compared, for example, to client equity loan because cost of risk on cash loan is higher than on culturalized lending, so it's basically -- it is seen that much, but it's more of IFRS 9 effect in the subsegment.
Next question comes from Andrzej Nowaczek from HSBC.
I have a couple of questions on fee income. First of all, I'm not sure I understood your answer on reclassifications in Q2. Maybe it is related to reclassification, but I'm looking at the balance between what you earn on interchange and acquiring commission and what you pay to credit card companies, and this is turning increasingly negative. Is it because it just takes more incentive to bring the customers in now? Are those customer acquisition rates sustainable? For how long?
Sorry, just to clarify. It can't be negative. But basically, maybe you're lumping different things together that shouldn't be lumped in. So we have -- as an acquirer, we earn income as an online acquirer predominantly, yes? We earn income from the merchants who pays a certain amount, and then we pay fees to the payment system, and we also pass interchange through the payment system to the issuing bank. And there, obviously, is positive. That shall be taken separately from our issuing business, where as an issuer, issuing cards, we're receiving interchange. I'm not sure we netted out against cash back that we're giving back, and maybe some other items that we used to net. So Ilya can take you through this, but those shouldn't be seen together because they're completely separate businesses. Ilya, maybe you should explain more detail, and then Andrzej can clarify his question.
Yes, well -- or maybe, if you basically compare the -- in the fee and commission income, you compare acquiring commission plus interchange fee to payment system commission expense. Then on the income side, you should also take cash withdrawal fee and card-to-card commission because there's basically the in-depth payment system line in the fee and commission expense, it's not just items related to basically the Mastercard and Visa. And then there is also certain deferral mechanisms, for example, cash withdrawal fee we defer over a longer time.
Just to add to that briefly, we've been doing a lot -- we mentioned this on previous calls, I think. We've been doing a lot in terms of promotional activity for lifestyle services. So for travels, cinema, theater, that kind of thing, concerts. And we've been doing different promotions with cash back. And so there is a bit of an increase in, well, significant increase in cash back paid in the first quarter going into second quarter, and that was netted out. So maybe that's something you're seeing, but there's certainly no negative numbers there. That may just be a slightly suppressed trend. Did we answer your question? Did we understand it correctly?
Yes. Yes. I noted what you said about treasury income as well. So presumably, you've done certain increase and should be added too, right? But the dynamic of turning slightly more negative, and I was thinking, okay, maybe it's just tougher to bring customers on board now although it's not the issue.
No, no, no. So I suspect that it's got to do with -- the answer really lies in cash back. So it's just the netting due to a lot of promotions we've done, but there's no negative dynamics there.
Sure, sure. And one other question, a quick one on fees. Why are credit protection fees not growing?
So Andrzej, you're talking about the credit protection insurance on our credit products. So basically, we now move more of them to the insurance business because some moment last year, we basically -- we changed the coverage. We increased the coverage on our -- on this part of insurance. For example, we included loss of job to the coverage, and therefore, we are now booking more on the insurance side. That is why it's not growing on the banking side.
We can now take our next question from Elena Tsareva from BCS Global Markets.
Just a bit of small clarification, more details on what you said about asset quality worsening, as you named, like the trends now. Just -- I understand you break down cost of risk and tried to sell. Now with this kind of -- can come from worsening, but it's also mainly like the discussions now around that there is a bubble in retail, lending, and we are now here, I think you're foresaying that there is asset worsening. Just maybe you can just say how it compares early stages of asset quality to the previous cycle when you see deterioration of asset quality? Maybe you can guide whatever could be a normalized cost of risk now for credit cards? I think before, previously, we expected that longer term, it could be around 9%, if I'm not mistaken. Maybe you can provide more details how you see in your client segments where you can see it worsening, and maybe you do some actions to prevent more worsening in the portfolio? And if you see a regulation tightening, it will help to maybe bid to stop this problem or actually will work in the station, given that many leveraged clients will just default because they're like, there will be no opportunity to refinance. If you just make color on all this issue, it'll be helpful.
Sure. Elena, I'll try and answer this. So lots of the different parts to your questions, I'll try and break it down to bite-sized chunks. So we say we're passed the midpoint in the cycle. So what does that mean? It means that from, basically, 2015 all the way through to the end of 2018, our cost of risk was coming down. It was mainly coming down for most banks in the market, but not for everybody, we're generally speaking. And this was a result of deleveraging of customers, particularly in '15 and '16. It was the fact, there was very little disbursement of new loans, and really, only the market started growing again at the beginning of 2017. It was changes -- you could call the structural changes in the market if you like, so different set of regulation. So it went from being unregulated to being very regulated. The consumers being through the watch as a borrower understands what the credit history is, and obviously, a different set of lenders. With the different -- with [ gray hair ], let's put it that way. So we go all the way through the end of 2018, which we think is -- marks probably about the plateau, the bottom of the cycle, and we see a softening of some numbers starting from the end of last year, beginning of this year. I'll comment on sales number in the second. But basically, we're seeing more lending being done. The rate of consumer lending has picked up, obviously through 2018 and 2019. You hear some remarks made by the -- by some government figures in Central Bank. They're worried about the potential appearance of a bubble or bubbles in consumer lending somewhere down the line, not just in unsecured but in secured lending as well. And others, obviously against the backdrop of people's income stagnating, if you take the average numbers. But as we pointed out many times in the past, there's different parts of the Russian population. You have to look at different associate demographics, geography, and you see, unfortunately, 2 Russias. One of which is doing better and the other one is doing worse.So you can see -- I should say that we've always said that our, through the cycle, cost of risk, we believe we'll be 10% to 12%. We now -- what we think is second half of the cycle and what that means to us, is that you will see the cost of risk either behaving stably or maybe starting to tick up. So that doesn't mean we're in a lending crisis or there's lending bubbles far from it, but it does mean that you will see a gradual deterioration of credit costs. And then the speed with which it deteriorates, determined by the external environment, obviously, macro and the economy; b, what the regulator does; c, what other competitors do, in particular, to our customers; and d, what we do as a lender. So our risk appetite and the speed with which we're lending and which credit products we're pushing.So coming back to asset quality, what's gone a little bit soft? Most of our -- we always talk about the same set of metrics, and so for consistency's sake, we'll refer to them. First payment default, second payment default tells you something about new vintages being booked, and they move around depending on the season, depending on the channel, depending on what we're doing. It went up a bit in credit cards as a result of lots of stuff that we've been doing, including tests, and now they're coming down again, but more or less, stable. It tells you something but it's entirely a huge amount. More importantly is delinquency rates. This is the speed at which the existing back book, the existing customers are going to delinquency. And there, within the range of 1% either plus or minus, we're pretty flat. So there's nothing moving in any particular direction in any of our major credit lending lines -- business lines. And then you have 2 numbers which have gone a little bit softer, roll rates. So as Ilya have mentioned in his presentation, when a customer goes into delinquency, they're becoming a little bit more difficult to collect. So that means the likelihood that they're going to migrate deeper into collections from 1 collection cycle, 1 to 30, 31 to 60, 61 to 90, et cetera, has increased. Not dramatically but it softened, and that said, particularly in the unsecured lending lines.And approval rate. So we've kind of documented this, over like the last few years, it's something we normally talked about on our calls with you. So back in 2016, 2017, in credit cards, for example, we had 30% to 35% approval rate. Now it's ticked down to 20%, 22%. It's probably going to tick down further. So what does this say? This says that the quality of the incoming application flow has somewhat deteriorated. So the number of applicants with a loan, currently delinquent with another institution with a high PTI or DTI, who will have a default or deep delinquency on a previous loan, just have a lot of loans outstanding in terms of numbers, their share in the incoming application flows increased, which means we're not approving them. So it's just as where we are in the cycle, mid-cycle just passed. What will be the effect of PTI on this? For example, if it's done too drastically, too quickly, then the -- let's say, a more worst case scenario for the market is obviously, this pushes some of the more marginal borrowers into the state of delinquency, and it would, if you like, be the catalyst for a mini consumer lending crisis. And the Central Bank and their wisdom, I'll say that without any interval at all, have actually come one with a benign set of PTI thresholds for the beginning. And then let's see whether they tighten up as we go along. We don't know where the Central Bank is coming with a, what I would call, a very reasonable level with a reasonable approach, which is not an outright prohibition and saying increase risk rates. And then they'll presumably monitor the situation, decide what to do as we go along. And so this should not be the cause or catalyst for any problems in the market. The market will correct itself, it will slow down over the next 6 months. And you'll see us slowing down, for a number of different reasons, partially as we reduce our approval rates, partially as we readjust some of our lending away from some secured lending types to -- sorry, and unsecured to secured. As we recalibrate our NPV models on building, factoring PTI into that. So there's lots of stuff going on, and there's some moving pieces, normally. So hopefully, that gives you more color. I don't know if I missed anything else in your initial question.
Just a quick follow-up. So it's like, for the cycle, cost of risk of 10% to 12%. This is for blended portfolio for all types, so like secured or unsecured or just for the credit cards only?
No, that's a good question because obviously, the mix are changing quite quickly. So let's say that's for unsecured, so obviously, the mixed number would be -- the blended number will be, we'd expect it to be lower, but we just don't have the data to be able to tell you by how much.
That concludes today's Q&A. I would now like to turn the call back at the host for any additional or closing remarks.
That's it. Thank you very much indeed for your time.
Thank you. That concludes today's conference. Thank you for your participation, ladies and gentlemen, you may now disconnect.