Qfin Holdings - Q4 2019
March 27, 2020
Transcript
Operator (participant)
Thank you or standing by, and welcome to the 360 Finance Fourth Quarter 2019 Earnings Conference Call. All participants are on a listen-only mode. There will be a presentation followed by a question-and-answer session. If you wish to ask a question, you must press the star key followed by the number one on your telephone keypad. I would now like to hand the conference over to Ms. Mandy Dong. Please go ahead.
Mandy Dong (Investor Relations Director)
Thank you. Hello everyone, and welcome to our Fourth Quarter and the Full Year 2019 Earnings Conference Call. Our results were issued earlier today and can be found on our IR website. Joining me today on the call are Mr. Haisheng Wu, our CEO and Director, Mr. Alex Xu, our CFO and Director, and Mr. Yan Zheng, our Vice President. Before we begin our prepared remarks, I would like to remind you of the company's safe harbor statements in connection with today's earnings call. Except for any historical information, the material discussed on this conference call may contain forward-looking statements. These statements are based on our current plans, estimates, and projections. Therefore, you should not place undue reliance on them. Forward-looking statements involve inherent risks and uncertainties. We caution that a number of important factors could cause actual results to differ materially from those contained in any forward-looking statement.
For more information about potential risks and uncertainties, please refer to the company's filings with the SEC in the registration statements. In addition, this call will also include a discussion of certain non-GAAP financial measures. Please refer to our earnings release, which contains a reconciliation of the non-GAAP measure to the most directly comparable GAAP measure. Finally, please note that unless otherwise stated, all figures mentioned during this conference call are in CNY. I will now turn the call over to our CEO, Mr. Haisheng Wu.
Haisheng Wu (CEO and Director)
Thank you, Mandy. Hello everyone. Thanks for joining our earnings call today. We delivered another quarter of strong results to finish the year at a high note. During the full year of 2019, our loan origination volume reached CNY 198.67 billion. Cumulative registered users reached 135 million at the end of 2019. Users with approved credit lines increased by 12.8 million throughout the year. Total revenue reached CNY 9.22 billion, which is a quite strong beat to our guidance at the beginning of 2019. During the fourth quarter, we continued to adopt a stable and prudent management approach, taking ongoing regulatory trends into consideration which will actively slow the pace of our business development. Despite the challenging environment, we still achieved a good quarter with loan origination volume reaching CNY 53.12 billion. The fourth quarter of 2019 was atypical for China's fintech industry.
A series of regulatory requirements were rolled out, all of which presented significant challenges for companies that are relatively late in maintaining compliance standards. The measures also affected multi-platform loan borrowers, which in some way affected our business. Fortunately, we quickly and successfully adapted to this challenging environment. I believe we mainly benefited from two factors. Firstly, we remained in close communication with regulators on a regular basis as we are one of the leading fintech platforms in China. Secondly, our main target market is prime borrowers with credit cards who are more resilient to economic downturns. As a result, this further strengthened our cooperation with institutional funding partners. Next, I would like to brief you on a few key business updates during the quarter.
In the aspect of regulatory compliance, we successfully passed a rigorous assessment conducted through a joint government campaign by the Cyberspace Administration of China, the Ministry of Industry and Information Technology, the Ministry of Public Security, and the State Administration for Market Regulation. This was a direct result of our strict standards for data collection and personal information protection. In addition, we took the initiative to apply for and were one of the first batch of select companies to have our Mobile App approved by the National Internet Finance Association. In January 2020, regulators released a draft version of the Interim Measures for the Administration of Internet Loans of Commercial Banks, which legitimized the loan facilitation business model and removed the geographic restrictions placed on online lending business.
We view these regulations as beneficial to the fintech industry, which will allow us to grow in a more healthy manner. In the aspect of customer acquisition, we proactively slowed borrower acquisition activity. We strategically improved approval standards and the customer acquisition costs of each new borrower. We focused more on precise user targeting contributions through cost reduction of borrower acquisition and LTV accretion. In the meantime, we devoted more resources towards consistent borrower operations and as a result, the contribution to loan origination volume during the fourth quarter increased to 81.9%. In the aspect of funding, despite the turbulent market, we continued to add our institutional funding partners. We now work with a total number of 81 in the fourth quarter, up from 24 last quarter, among which nearly 40 operate on a nationwide scale.
While the geographic restrictions on online lending business have been removed, our internal stress test indicates that we are capable of coping with the most rigorous regulations. Our funding costs showed a downtrend momentum throughout 2019, falling by more than 100 basis points, and we expect this trend to continue in 2020. During the COVID-19 outbreak in the fourth quarter of 2020, we successfully issued CNY 500 million ABS as the fourth consumer finance ABS in 2020. In terms of equity stakeholders, we feel honored to welcome the world-renowned private equity group FountainVest Partners to become a significant strategic investor and board member in November last year. We were subsequently joined by our Chairman, Mr. Zhou Hongyi, and all key members of Management Team in December to announce a plan to jointly invest up to $15 million to purchase 360 Finance shares in the next 12 months.
By the end of December 2019, the Management Team and FountainVest Partners have already purchased $20 million worth of shares, in addition to the initial investment made by FountainVest Partners in November. FountainVest Partners' strategic investment and repurchase of shares by Management not only eased the pressure on our stock price due to lack of shareholders, but also demonstrated a solid source of confidence by Management in the future growth and development of our business. I'm very glad to share a couple of resolutions that were passed at our board meeting earlier this week. First, in order to further strengthen synergies with 360 Group, the board appointed Liu Wei, Senior Vice President at 360 Group, as Vice Chairman of our board. Second, one of our affiliates established an internet insurance business last year, which delivered very outstanding results during the COVID-19 outbreak.
The board authorized us to undertake a minority equity investment in this business, through which we expect to generate synergies with our loan business. I understand everyone's concerns about the impact of the COVID-19 outbreak on our operations and what measures we have undertaken in response. I would like to state with full confidence that, taking into account our 2020 year-to-date operational performance, our loan volume and risk management have already fully recovered after a short challenging period. This is another strong demonstration of our business resilience, attributing to our customer base and risk management capabilities. I want to highlight that we are one of the very first companies to take proactive action in response to the outbreak.
Before the widespread outbreak, we delivered loan collection devices to our staff at home and put in place setups to ensure the quick recovery of our business operations while working remotely in anticipation of a difficult operational environment following the Chinese New Year holiday. By the end of February, our loan collection operations have been 100% recovered. In addition, we and 360 Group jointly organized a number of charitable activities to donate and purchase medical supplies. During the Chinese New Year holiday, a team consisting of several hundred of our staff continued working to ship medical supplies from the world abroad to Wuhan. This reveals that we are not only a team with professional capability but also with a strong commitment to corporate social responsibility. Now, a few words on our 2020 outlook. The largest uncertainty comes from the ongoing COVID-19 outbreak, especially overseas.
Hence, we will implement a more prudent approach to manage risk and wait for opportunities. In addition, I believe platforms lacking essential key elements to operating a fintech business will be impacted to a much larger extent than we will be. Our affiliation with 360 Group provides us with structural advantages in terms of compliance, funding, risk management, and customer acquisition, which will allow us to take more market share in the long run. Taking into account our 2020 year-to-date operational performance, we are confident to achieve our business goals in 2020. Now, I will turn the call over to Alex to discuss our financial results.
Alex Xu (CFO and Director)
Thank you, Haisheng. Hello everyone. Firstly, let me give you a quick update on our full-year financials. Our total net revenue reached CNY 9.22 billion in 2019, a remarkable 107% year-over-year increase, and non-GAAP net income reached CNY 2.75 billion in 2019, a 53% year-over-year increase as well. We are thrilled to deliver this result, and this is a strong beat to our guidance: CNY 8 billion-CNY 8.5 billion net revenue announced in early 2019. This solid result further proves our capability to fully deliver our promises committed to all our stakeholders. Specifically, in the fourth quarter, as Haisheng just mentioned, our stable operations translated into healthy financial results. Total net revenues increased by 53% year-over-year to CNY 2.4 billion, and non-GAAP net income reached CNY 515 million in the same quarter.
In the face of the industry turbulence during the fourth quarter, we proactively adopted a more prudent strategy and successfully carried out a series of initiatives on various operational fronts, such as risk management, borrowers' acquisition, fund management, etc. In summary, we focused on two aspects: one, the increase of the operational efficiency, and two, maintaining sufficient margin of safety. In terms of operational efficiency, we witnessed a significant reduction of borrower acquisition costs, continued the decreasing trend of funding costs, and enjoyed a full-year low effective tax rate. Firstly, we proactively refined borrower acquisition strategy and trimmed down sales and marketing expenses. The acquisition costs continued to drop to CNY 228 for each new borrower with a full credit line, in comparison of CNY 246 in the previous quarter.
With this, we delivered the decrease in unit acquisition costs in two consecutive quarters, and we are confident to see the trend continue in the first quarter of 2020. Secondly, in spite of the challenging industry environment, we successfully issued CNY 1 billion ABS with an attractive all-in cost at around 5.6%. By this, our total ABS issue for the full year of 2019 reached CNY 2.3 billion, with an all-in cost of 5.6%. Moreover, we have expanded our cooperation with financial institutions. The number of financial institutions working with us increased to 81 by the end of 2019, in comparison to 26 by the end of 2018. Thanks to all these efforts, our overall funding costs slowed down from 9.2% in the fourth quarter of 2018 to 8% in the fourth quarter of 2019.
Thirdly, we have spent a great effort to reach a record low effective tax rate of 14.5% in 2019, in comparison to slightly higher than 20% in 2018. Again, this is a good indicator to show our sound relationship with all the regulators. In terms of margin of safety, we focused on maintaining a healthier leverage ratio and we proactively increased our provisions. Firstly, we continue to expand the capital-light model business, which is without any risk-taking. During the quarter, loan origination under the capital-light model accounted for 22% of the total loan origination volume, up from 20% in the prior quarter, with a Q-on-Q increase of 27% in terms of outstanding loan balance. While our total outstanding loan balance continued to grow, the outstanding balance with risk-taking business decreased to CNY 58 billion from CNY 59 billion last quarter.
Our net equity increased by 6.9% year-over-year to CNY 7.2 billion, as our leverage ratio continued to decrease to 8.1x from 8.8x in the previous quarter. We expect that leverage ratio will continue to its downward trend in the coming quarters. Secondly, in terms of the provisions, we would like to highlight that different platforms might undertake different accounting approaches, and as a result, there may not be a direct like-for-like benchmarking analysis among different platforms. Our approach is to assess credit security in the perspective of the entire loan life cycle and book a sufficient provision covering the whole cycle at the inception of the loan origination. We add more provision this time when we see the negative impact is slightly worse than our original expectation, so as to maintain 4x provision coverage.
More importantly, before loans expire, we do not recognize any gains from over-reserved provision items when asset quality turns out to be better than our original expectation. Therefore, you won't see any negative numbers on any of the provision items on our P&L. The gains from excessive provision will be booked under other revenue lines only after loans retire. In light of economic uncertainty and impact on the collection process, we enhance our provision cushion for the whole outstanding loan portfolios and maintain a sufficient coverage ratio of more than 4x. Take into consideration our 2020 year-to-date operational performance. We preliminarily expect a further increase of provision due to the impact of the COVID-19 situation, but we can assure you this will be a manageable level.
Furthermore, in regard to compliance, I would like to add that the contribution from our related P2P funding continued to decline in the fourth quarter. As of December 31, 2019, the P2P funding dropped to 4.5% of our total loan outstanding balance. We expect that this downtrend continues in the coming quarters. Currently, the credit line granted by our institutional funding partners is more than 2x our total loan origination volume, which provides easy replacement for the P2P funding. This is another solid demonstration of our sufficient and varied institutional funding resources. In closing, despite the unexpected outbreak of COVID-19 and the volatile global economic conditions, our business remained stable and healthy.
In the meantime, we will continue to focus on asset quality in line with our margin of safety through careful cash management and also stay alert on any minor turmoil down the road aiming to accomplish our full-year business target in 2020. Additionally, I would like to emphasize that our management are in a firm consensus that we are facing a golden opportunity to further solidify our market-leading position and remain with full confidence in our long-term growth prospects. Now, I would like to turn the stage to our CRO, Yan Zheng.
Yan Zheng (CRO)
Thank you, Alex. Now, let me give an update on risk management. We maintained a fairly stable risk performance through 2019. In the face of industry turbulence during the fourth quarter, we undertook a more prudent risk-staged strategy to keep our asset quality. We adopted this prudent strategy through the whole product life cycle. To elaborate in more detail, when we granted credit lines to customers, we adopted a tight credit approval policy for new customers, leveraging a robust risk model with higher precision and more underlying variables. After borrowers made a credit drawdown, we enhanced the management of customers with relatively worse credit profiles in order to control risk exposure. Our day-to-day delinquency ratio was 6.77%, with a modest drop compared to 6.93% in the last quarter of 2018, and a slight increase compared to 6.4% in the third quarter of 2019, despite the turbulence environment.
This is a clear demonstration that our risk policy refinement came into good effect already. In addition, during the second quarter of 2019, we started a pilot program, applying different terms to borrowers according to their credit profiles. In the A/B test, we offered different products so that borrowers with relatively bad credit profiles were provided with longer terms to ease financial pressure, and borrowers with relatively worse credit profiles were provided with short terms to accelerate their repayments. In the fourth quarter, we took more efforts to shorten loan tenors to borrowers with worse credit profiles and intended to scale up this pilot program in the future. In the loan collection process, we continue to improve the performance of AI robots and increase effective communication times of our collection teams through technology refinement.
In addition, I would like to highlight our system automatically logs every collection action, including activities of outsourced teams. 100% of the records are monitored in real time by experts and ensure compliance to the greatest extent. In the face of the coronavirus outbreak, we continue to maintain this more prudent risk management strategy as we did in the fourth quarter. We quickly tightened credit policy in those areas mostly affected by the pandemic and for those customers with occupations that may be significantly affected in the economic slowdown. Nevertheless, we still understood some pressure of asset quality during the outbreak. We expect our day-to-day delinquency during the first quarter to be in the range between 7.2%-7.3%. As the outbreak gradually stabilized in China, we saw day-to-day delinquency and collection rates have normalized and improved gradually.
Moreover, we took rapid and effective management of both in-house and outsourced collection employees. As a result, collection operations were 100% recovered by the end of February. During the outbreak, we are glad to find out that borrowers in some certain risk categories stay at their normal risk level showing strong risk resilience. All of those data during the outbreak, which is a huge stress test, are valuable to further refine our customer acquisition strategy and risk management strategy. We will use the data to further identify our borrowers according to different credit profiles and explore more differentiated risk strategies. In 2020, we have continued to implement this competitive strategy and have full confidence in maintaining the risk management excellence. That's all. Thank you. Alright, we can now start the Q&A session. Thank you.
Operator (participant)
Thank you. If you wish to ask a question, please press star one on your telephone and wait for your name to be announced. If you wish to cancel your request, please press star, then two. If you're on a speakerphone, please pick up the handset to ask your question. We will pause momentarily to assemble our roster. The first question today comes from Jacky Zuo. Please go ahead.
Jacky Zuo (Research Division Analyst)
Hi, good afternoon. Thanks for taking my questions. I have two questions. First is on the loan originations and our guidance. So can you give us some color on the first quarter loan originations, given that we are coming to the end of March? And also regarding to our 2020 loan origination guidance, I just want to know what the assumptions behind, for example, what's the distribution of the quarterly originations we are expecting for this year? And second question is about the asset quality. So it's very helpful to know our day-to-day delinquency rate is expected to be 7.2%-7.3% at the end of this quarter, up from about 7, sorry, 6.8% from last quarter. So I just want to know in terms of vintage loss, what will be the level we're expecting given the current virus?
And also, I noted that we actually set aside a large provision during the quarter. It's over CNY 700 million. So just want to know what is the assumption behind this additional provisioning? It seems it's about 1% of our loan balance at the end of last year. Is that the kind of level of provisions we expect for the impact from the virus? Thank you so much.
Alex Xu (CFO and Director)
Thank you, Jackie. Let me take the question first, and probably Haisheng and Jianyang will add on that. This is a very important question. The first one is the guidance. Yes, we issued our guidance for the full year's loan originations, CNY 200 billion-CNY 220 billion. The rationale behind that is that standing at today's situation, it's a little bit unclear for the future, but we are very confident that everything is under control. But just as you know, our companies always take a relatively conservative measure in terms of daily operation. So we want to at least keep the guidance as flat as last year because last year, the overall loan origination, the total loan origination is CNY 198 billion. So that's the key rationale for that.
Hopefully, if we see any recovery signs from the outbreak of COVID-19, this number will have a positive adjustment in the future. Just to add some color on this underlying assumption is we would expect the capital-light model will continue to grow to fulfill our strategies that technology enables. The second thing is that we will continue to see the acquisition cost to remain at a relatively low level. That's the information we can share with you on this call. Your second question talking about the asset quality? Oh, yeah. For the first question you asked about Q1, as of today, all the numbers are subject to the auditor's review, but all we can say is that we see the loan origination for the first quarter should be around CNY 50 billion.
So that gives you a rough sense of how we might conclude the guidance for the whole year. Your second question talking about asset quality, yes, I think I will defer the day-to-day delinquency ratio question to our CRO. But for the provision, let me explain a little bit further. As I mentioned in my speech just now, different platforms have different measures to, I would say, accounting approach to take provisions. Our approach is to assess asset quality for the entire loan life based on the situation we were facing. Okay? So what we do is to, standing at the Q4, the situation, we reevaluate as usual for the whole historical assets or outstanding balance. We will see that there is some impact on their asset quality. And so we take the additional provision.
In terms of vintage, what we can see is by the end of Q4, the overall asset vintage is around 2.6%. It's still manageable. That number can be compared with maybe our CRO will mention later the full year estimate for this year. I will defer the day-to-day delinquency ratio question to our CRO.
Yan Zheng (CRO)
[Foreign Language] 好的，我先说中文。说完之后，我们同事会帮我翻译。在整个2020年全年交易，我们预估我们的day-to-day delinquency会在6.5%-7.5%之间。然后在我们对这些交易在完整表现之后的生命周期的loss，也就是vintage的delinquency这一块，我们预估全年会在2.5%，平均会在2.5%-3.5%之间。但这个还取决于说接下来的疫情是否有可能会反攻。但目前来说，我们看到国内的疫情控制的趋势控制非常好，而且我们的各项风险指标也稳定得非常好，并且逐渐在改善。我们非常相信中国政府，并且很有信心控制把我们刚刚说的这些指标控制在这个范围之内。
Okay. Thank you, Mr. Zheng. So for the new transactions in 2020, the day-to-day delinquency rate will be expected to be around 6.5%-7.5%. Our expectation of the vintage delinquency rate will be around 2.5%-3.5% on average, which depends on if the ongoing COVID-19 virus outbreak in China will be eased soon or not. For now, we can see that the epidemic control in China is very effective. We believe in our government, and we are fully confident that our credit performance will be fully within our range of expectations. Hope that can clarify.
Jacky Zuo (Research Division Analyst)
Thank you. It's very clear.
Operator (participant)
The next question comes from Daphne Poon with Citi. Please go ahead.
Daphne Poon (VP and Equity Research Analyst)
Hi. Thanks for taking my question. So I also have some more questions regarding your provisioning and your asset quality. So first is we see that in the fourth year, on-balance sheet provision actually also increased. And if we calculate the annualized credit cost, it's around 13% of your average on-balance sheet loan balance. So it just seems like quite a high percentage. Just wondering whether that is also because of front-loading or sufficiently factoring in the coronavirus impact or that effect just because of the Q4 rising asset quality risk? And also, do you see on-balance sheet loans have a worse asset quality compared to the off-balance sheet ones? And then the next is want to understand more about your provisioning outlook for Q1. In particular, I think on the guaranteed liability, there's a big number in Q4.
So just want to see, just want to confirm whether that has also taken into consideration of the coronavirus impact and how much more provision you expect to set aside in the first quarter. Do you expect a similar around CNY 700 million additional guaranteed liability? So should we see that as a recurring number for Q1? And also regarding your guidance, so just want to confirm because it just seems that you mentioned earlier your Q1 loan origination is already at CNY 50 billion and your full year is around CNY 200 billion-CNY 220 billion. So it just seems very conservative given you also mentioned you feel pretty confident about the recovery of your asset quality and coronavirus situation. So just want to understand more on the rationale behind.
Do you see the whole industry that the penetration is high and the room for growth is declining so that you turn more conservative on the customer acquisition, or is it just because of the uncertainty of this macro or coronavirus outbreak situation here? Lastly, it's on the capital-light model. I wonder if you have any specific targets or indicators of ramp-up of that model and what percentage of contribution do you expect to have, say, by the end of this year? Thank you.
Alex Xu (CFO and Director)
Sure. Thank you, Daphne. For your first question, the provision on the balance sheet assets, what we can see is that we treat on-balance sheet assets and off-balance sheet assets in the same way in terms of taking provisions. The methodology is the same. The only difference is that the asset quality might vary on different types of under different types of models. In terms of methodology of provision, it is the same. Okay? To answer your second question, just following the accounting rules, we are not allowed to take in provision for the things that never happened in that quarter. For example, in the fourth quarter, if we take a provision, strictly follow the accounting rules, we can only take provision based on the assumption that there was no coronavirus. That's the base assumption.
But as you know, this is just an estimate, and we tend to provide higher provision coverage. So first of all, we follow the rules, accounting rules. Secondly, we put our best effort to do the estimate. So in the Q1, though it's already 27th of March, but in terms of the provisions, we are still in discussion with our auditor. And we can't give you a specific guidance whether this number will what number it will be. But what we can reemphasize here is that numbers should be manageable. Okay. Question 3, the guidance, the conservative guidance is not based on our forecast for the whole industry or say the outlook of the industry, but purely based on the uncertainty of this COVID-19 incident. Because even China is under control, the whole world is in turmoil.
So until those factors fully show up their impact, we will take a relatively conservative in terms of guidance. But as you mentioned, we will closely monitor the whole daily operation, all the data to see if there's any signs of recovery or certainty in the future in this year. We will definitely speed up our growth in the coming few quarters. So that's the reason we are very confident to deliver this guidance. The fourth question is about the capital-light target model. We do have a general target for the capital-light proportion. By end of last year, 2019, the new loan origination under capital-light model contributed around 22% of the whole book. And we expect that number will increase to 30%-40% by end of this year under this relatively conservative guidance. Hopefully, I answered your questions. [Foreign language]就中国持有，因为我们的用户人群里面最典型的最大部分特征就是持有信用卡的人群，这也是我们一直在主流经营的。中国持有信用卡的也有4.2亿人，然后其中信息资产就是接近50%，也就是说有一半人是愿意去有偿地去从信用卡借钱的。所以这些人都是有可能是我们的潜在目标用户。也就是说有一半人的话，4.2亿乘以50%左右就是大约有2亿人。2亿人是在信用卡里面有信息的资产余额的。我们今天在在的用户其实也就1000万左右。所以其实潜在的看市场规模空间的话，我们还有20倍以上的移动空间。所以这个问题远远不是一个市场容量的问题。谢谢。
Yan Zheng (CRO)
We have Alex's question, and our CEO, President Haisheng, also says that actually there are like 420 million customers, the credit card holders in the China market, and around that 50% revolvers that are willing to borrow money from the bank. So these are our potential customers. And compared to what we have now, that means that 20 times the customer space is still available for us to grow. So to add on Alex's answer of that question.
Steven Chan (Executive Director of Equity Research)
Yes. Thank you. So I just want to follow up on the capital-light. So you mentioned 30%-40% is based on the con ervative loan volume guidance. So does this mean that the percentage will drop if the loan volume turns out to be better than your current expectation?
Alex Xu (CFO and Director)
Well, we will evaluate that percentage. If the outlook for this year is better than we expect today, then we will give the guidance accordingly.
Daphne Poon (VP and Equity Research Analyst)
Okay. Sure.
Operator (participant)
The next question comes from Steven Chan with Haitong International. Please go ahead.
Steven Chan (Executive Director of Equity Research)
Good evening, Management. I think a quick one, if I can. One follow-up question on the loan origination volume guidance. And you talk about, I think you have mentioned much about all the assumptions you based on. But I would like to know one key assumption. What will be your assumptions of the increase in cumulative borrower in order to achieve the loan origination guidance for 2020? That's the first question. And secondly, you mentioned that you're going to increase the share of capital-light model business.
I would like to know, because you have increased your provision coverage for the non-guarantee, so does that affect your effective return or take rate from your non-guarantee model, meaning that if you are making more provisions for the guarantee model, does it imply that for the non-guarantee model, well, the return you can get also have also been revealed starting in the final quarter last year? And sub-question for that is, do you have any target of having 100% of your loan origination being capitalized? So that's the second question. And finally, I think a follow-up on these new items called guarantee liabilities. I would like to see for your clarification again. So is this item related to preparation for the new accounting rules in 2020?
So that's why you're trying to make more provisions now and make sure that that will likely reduce the potential negative impact, especially in the first half of 2020, if the economy or unemployment rate or whatever, delinquency is very, very high. And sub-question for that is, in case if the economy starts to pick up very significantly in the second half of 2020, are you going to see potential rise back? Because it seems to me that it's very similar to the IFRS 9 for banks. So when the economy is recovering, probably the provisions will likely be reduced. So I'm not sure whether you are trying to make a lot of excess provision in the final quarter of 2019, and then that could be too much, and probably that could be challenged by that if the economy actually picked up very fast.
Alex Xu (CFO and Director)
Okay. Thanks, Stephen. Very good question. The first one, yes, we do have some underlying assumption in terms of new borrowers to generate this guidance, to meet the guidance. And under that assumption is we will continue our strategy in the fourth quarter last year, i.e., we focus on our existing borrowers who we know much better instead of acquiring more new customers under such circumstances. So based on that underlying assumption, our new borrowers for 2020 under such a guidance will be much lower compared with last year. And we can easily meet up the guidance, primarily focus on our existing customers and to better serve them. The second question in terms of capital-light, a quick answer for your second piece of your question is ultimately this capital-light model will contribute the majority part of our business in the long run, but not this year.
So 100%, even if it will be accomplished, it will be in the at least five years horizon. Okay? The second part is your first piece of your question is the return on the capital light, whether they have any negative impact triggered by the asset quality decrease. In short answer, no. We have maintained our sound relationship with our financial institutions because in the commercial terms with those financial institutions, the threshold to decrease our service charge is very high. Of course, it's based on the industry general practice and industry standard. But just because our historical vintage is so much better compared with the industry average, so we have a significant buffer in terms of commercial terms. So even in February this year, we still see strong support from the financial institution under the capital light model as well.
So to answer your question, there is no negative impact on that front. We still have the similar unit economic aspect to it. The third question about the expense of guarantee liability. As I mentioned, this is Q4 results. So we didn't adopt a new accounting policy in the fourth quarter. The impact of that accounting policy is still in discussion with our auditor. But based on our preliminary discussion, we don't expect a very negative impact on us. The influence will be very limited. That's the only thing I can mention at this moment for the ASC 326. And the other thing is you mentioned how to when the loan expires, the excess coverage of the provision will definitely come back to our revenue. We did that every single quarter.
Even in the fourth quarter, we have more than CNY 70 million come back from the guaranteed liability to our revenue item. So we will see that trend continue this year and going forward. Hope I answered your questions.
Steven Chan (Executive Director of Equity Research)
Can I follow up on that just for the first question? So under your guidance on the new cumulative borrowers, so to put it simple, there could be high chance that we're going to see a decline in sales and marketing expenses in 2020 compared to 2019?
Alex Xu (CFO and Director)
Yes. Yes. Yes. Yes.
Steven Chan (Executive Director of Equity Research)
Thanks.
Operator (participant)
Once again, if you have a question, please press star then one on your telephone and wait for your name to be announced. The next question comes from John Cai with Morgan Stanley. Please go ahead.
John Cai (VP)
Hi. Good evening, Management. Thank you for taking my questions. I have question three. So the first one is just to see if the management can share with us some insights on how they assess the current credit cycle. Is it fair to say that the worst is over? And what metrics would the management monitor to see whether we can accelerate the growth or not? And related to that is we have heard some banks are getting very prudent in terms of the credit card business. And some of them might reduce some credit lines, and they see some risk increase in the portfolio as well. So if the banks scale back their credit card business, do we see it as a risk factor or an opportunity? And the reason is because we have our borrowers overlap with the banks' credit card customers seems to be high.
So that's the first question on the industry. And the second question is also related to the guarantee and the 4 times coverage. So just wonder what's the 4 times mean? Does it mean our reserve balance divided by the delinquent loan is 4 times, or if there's any colors, we can share on that metric? And so the final question is about basically customer acquisition and cash because on the customer acquisition, we have seen both the total number decline in the fourth quarter and also the unit cost. So it's a very good trend. And just wonder going forward, do we expect the customer acquisition cost to remain at this low level? And is it a function of our quantity and the unit price? That means if we increase the volume, the unit price will go up.
So if we keep it lower, the unit cost is also low. And then so in the fourth quarter, we reduced the asset marketing expense. And since our loan balance is going much slower, so if we keep the acquisition cost low, do we see any pressure to grow the loan balance? And with our strategy to transition to the capital-light model and reduce or lower the sales anN marketing expenses, since we will be building up cash in 2020, I just wonder if there's any other plans on how we use the cash on hand. Thank you very much.
Alex Xu (CFO and Director)
Thank you, John. I just put out a list of 4-5 questions. Let me answer some things first, and then I will leave the record to my CEO and the CRO, Zheng Yan. And just backwards, okay? For the customer acquisition, yes, as I mentioned in my speech just now, we do expect the acquisition costs will follow this decreasing trend in the coming few quarters and maintain at a relatively low level. That said, this is based on our conservative guidance at this moment. Okay? So to some extent, you are right. If the economy recovers quickly in this year, we might decide to speed up our growth, and we will invest more on customer acquisition in the coming few quarters.
And that unit cost might go up a little bit, but it's still too early to say because we take a lot of initiative to fine-tune our customer acquisition strategy and focus more on the high stickiness and high quality customers in the coming few quarters. And we will see the positive impact on this acquisition strategy as well. Yes, if the situation continues like this or even getting worse, of course, we will have more cash than the normal business needs. We might consider the cash for some capital market activities. We don't rule out the options. Also, we might take a closer look at the insurance business that we just plan to take a minority stake in as announced just now by Haisheng. But this is just options. We will carefully evaluate the situation based on the overall macro and micro data we can get. Okay?
That's the customer acquisition question. The third one is the guarantee liabilities fair value. Yes, this is a very good chart. Thank you for giving me an opportunity to explain this. Okay? As I mentioned, different platforms have different policies on that, but we maintain our prudent strategy to give a high cushion to our delinquencies. So just put it in a very simple way might not be absolutely correct. You can take a look at our M3+ delinquency ratio, then times 4, slightly higher than 4. That will be our provisions, the total provisions for the loans, all outstanding loan balances. And the second question regarding the banks' scale back of credit card, if that is an opportunity or not or a risk for us, I think I will leave that question to our CEO. For the credit cycle, the macro credit cycle question, I will leave to our CRO.
Yan Zheng (CRO)
我先回答一下说 credit cycle 这个问题。实际上我们在过去的疫情中，我们看到说不同的客户在这次疫情是受了不同的影响的。比如说我们看到湖北省的风险，它在最高的时间点是之前的两倍左右。但是它占我们总资产5%不到，所以影响不大。现在也慢慢在回落。然后不同的职业受影响也不太一样。比如说像文娱职业、酒店职业和零售业的话，这风险会有一个相对比较明显的上升。而金融、政府、教育和医疗行业的风险则基本没有影响。我们会预计说像制造业，特别是跟海外那边有关的一些批量制造业，可能在未来会受一定程度的影响。像快递行业，我们就看到说它的风险在经历一个非常短暂的略微上升之后，快速就回到了正常水平。对。所以我们对这些客群做了非常多维度的聚类分析之后，发现说实际上我们有80%的客户在这次疫情中风险是完全不受影响的，甚至有一小部分客户他们的风险甚至还低于之前。我们也有15%的客户是受轻微的影响，还有5%的客户受了比较严重的影响。所以我觉得这个数据其实对我们来说是一个很宝贵的财产。我们根据这样的，其实这次疫情我们会把它当成是一个类似于经济危机的这么一个事情。根据这些信息，我们后面会针对性地去对一些这种受经济影响比较小的客户去做更深度的经营，来减少说那些可能会受更大影响的这种客群。对。然后关于这些指标的监控，我们是有一套很严密的指标，包括早期的一些预警的风险指标去监控这些信息。我们也有一套模型去监控。我们有一套不变的，就参数不变的一个模型去监控说不同时期这些客户同一模型分数下客户的变化情况，这个可以帮我们在一定程度上去预测一个现在的情况变化。要不请我同事翻译一下。
Okay. So our CRO raised the pandemic situation now in China as an example to answer the question of the credit cycle. So actually, we have discovered that the pandemic has different impacts across different customer groups on our assets. For example, for Hubei, the highest delinquency rate during this period has doubled, but it accounts for only around 5% of our total assets. And for different industries, the impact varied as well. So for instance, sports, fitness, and entertainment industries, hotel and catering industries, and retail industry have been affected a lot. While financial services, government education, and medical services have not been affected. While manufacturing, especially those who have orders from overseas countries, might be affected later because of just arriving overseas countries, while delivery industry increased at the beginning or normalized very fast.
So with all this data, we have done multi-dimension clustering analysis, and we found that around 80% of our customers are not being affected at all or even have lower credit risks, while 15% slightly being affected and around 5% severely disrupted. So those kinds of data and behavior are like a treasure for us and a real situation pressure test. So in the future, in terms of a credit cycle, we will emphasize on good customers that we have data and we have the evidence and the credit performance as well. And we also have quite stable indicators and models to monitor the changes of credit performance of the same customer. So we can see that we are fully prepared with all the data for the credit cycle. Hope this can clarify. So maybe the next question is for our CEO about the credit card.
Haisheng Wu (CEO and Director)
[Foreign language] 好。关于就是说银行的对新银行采取审慎的策略，对我们是危机。我认为在目前这个阶段，因为现在疫情带来的不确定性并没有消除，所以在眼前的阶段，我们不能认为它是一个机会。但是在疫情之后，我认为有几个方面的信息可以分享。就是第一是我们跟新银行之间的关系，我们有肆意的存在用户之间，我们是互相补充的关系。因为银行对于持卡用户的服务并不是非常的完美，所以才有了我们这种金融科技行业的存在的必要，我们让用户体验更好。第二就是我们其实也通过互联网的大数据，让银行未能服务的实际潜在用户也可能会被我们覆盖到。所以从这个角度，在市场确定性更高的时候，机会会更多一点。但前提是等确定性会小一些的时候，不确定性。对啊。
Okay. For now, with the uncertainty of the COVID-19 situation, we cannot just say whether it's an opportunity or it's not. In the future, when the situation is eased, and we can see that basically we are complementary compared with a credit card instead of a competitor directly, with the data and with all the credit card holders' space, we can have more opportunities when the situation is more certain.
John Cai (VP)
好，非常感谢。 Thank you very much.
Operator (participant)
The next question comes from Claire from Golden Dragon Asset Management. Please go ahead. Claire, your line is up. Claire?
Speaker 9
H. Am I heard?Yeah.
Thank you. So I have three questions. First, management mentioned that you are much more conservative compared to your peers. So can you give a bit more color on our guaranteed liability reserve ratio, which I guess is calculated as current quarter new provision of guaranteed liability divided by the amount of newly originated off-balance loan covered by guarantee? And my second question is, so how do we manage the collection of overdue loan during the coronavirus? And why are we able to manage our NPL kind of better than our peers? And my third question is also on the guaranteed liability expense. Can management give us more color on what's the definition of guaranteed liability expense? And how much of our expenses relate to the current quarter, and how much is more related to our expectation about the future and the full year? What's our guidance of this item in Q1 and full year? Thank you.
Alex Xu (CFO and Director)
Thank you, Claire. Let me take our guaranteed liability questions first, and I will defer the collection question to our CRO. So for the expense of the guaranteed liability, to put it in a so-called unprofessional way, it's more like a fair value reevaluation of the guaranteed liabilities. Okay? So basically, when the transaction happens, based on according to the U.S. GAAP, we need to estimate what is the potential loan under that transaction. So that number will go into the so-called guaranteed liability. So every single quarter, when we release the quarter report, before that, we need to ask a third-party, independent third-party to reevaluate all the outstanding balance in terms of to see whether the original expectation or the estimate is in line with the current situation. If the situation is worse than our original expectation, we will add more guaranteed liability. We'll add more provisions as well.
If the situation is better than our original expectation, we will do nothing under our current treatment. We will wait for the asset to expire, and the excess guarantee liabilities will go back to our revenue. That's the key methodology. This fair value change of the guarantee liability or, say, the expense of the guarantee liability is more like reevaluate all the outstanding balance in that specific quarter, and we will see what was their lifetime performance is like. We have our best estimate at that quarter for the future. We book that into this fair value change. This change under the accounting, based on the accounting rules we take, is purely standing at the end of Q4. We made a forecast for the asset quality.
We will reevaluate the asset quality by end of first quarter before we released the first quarter results. We will process it again under the same procedure. Hope I answered the guaranteed liability question. Then I will turn to our CRO to respond in response to your collection questions.
Yan Zheng (CRO)
Yeah. Sorry, Claire, can you please repeat the question about the collection actions in place during the COVID-19 situation?
Speaker 9
Yeah. So before that, can I repeat my other unanswered question? So first, we recorded more than CNY 700 million guaranteed liability expense, and we have CNY 71 billion outstanding loan balance. So does that mean that we are expecting NPL to increase around 1% because of coronavirus? And also, you also mentioned that our company is much more conservative than others in terms of reserving. So can you share with us what's the guaranteed liability reserve ratio for Q4 and also for the full year guidance?
Yan Zheng (CRO)
[Foreign language] 好的。那个我回答一下这个问题。首先在疫情期间，一开始我们因为我们能我们是在大年初三的时候开始逐步去把我们的电脑寄给我们的员工，然后我们已经开始紧急去联系我们的外包公司，让他给我们提供更多的作息。那与此同时，我们也把我们的机器人正好我们在四季度做了一些更新迭代的东西。那正好机器人在过年期间就用得非常非常的足。那在2月底的时候，我们的所有的外包人员和自营人员都已经准备到位。所以那时候我们就开始100%的就按照我们正常的管理方式在管理。另外就我们在疫情期间如果是碰到就疫情高发期高发的地区的那些疫区人员，我们跟银行其实有讨论过，跟我们资金方有协商过，就是我们会对他进行一些缓推的一些措施。但这些在我们的算法里都是算在已经逾期的，只是我们现在缓推他出于那个出于一些对客户的一个人道主义关怀。
好的。 Okay. So actually I'm going to ask you something. I just want to ask you something.
Operator (participant)
Okay.
Just to ask.
Yes.
Just to ask.
Okay.
Just to ask.
Yes.
Just to ask.
Yes.
Just to ask.
Yes.
Just to ask.
Yes.
Just to ask.
Yes.
Haisheng Wu (CEO and Director)
Yes.
Yes.
Operator (participant)
Yes.
Yes.
Yes.
Yes.
Yes.
Yes.
Yes.
Yes.
Okay. Yes.
Yes.
Yes. Yes.
Yes.
Yes. Yes.
Yes.
Yes. Yes.
Yes.
Yes. Yes.
Yes.
Yes. Yes.
Yes.
Yes. Yes. Yes.
Yes.
Yes. Yes. Yes. Yes.
Yes.
Yes. Yes. Yes. Yes. Yes. Yes. Yes. Mm-hmm.
Yes.
Yes. Yes. Yes. Yes. Yes. Yes. Yes. Mm-hmm.
Yes.
Yes. Yes. Yes. Yes. Yes. Yes. Yes. Mm-hmm.
Yes.
Yes. Yes. Yes. Yes. Yes. Yes. Yes. Mm-hmm. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes.
Mandy Dong (Investor Relations Director)
Yes.
Operator (participant)
Yes. Yes. Yes. Yes. Yes. Yes. Yes.
Yes.
Yes. Yes. Yes. Yes. Yes. Yes. Yes. 嗯。
Yes.
Yes. Yes. Yes. Yes. 嗯。
Yes.
Yes. Yes. Yes. Yes. Yes. Yes. Yes. 嗯。
Yes.
Yes. Yes. Yes. Yes. Yes. Yes. Yes. 嗯。
Yes.
Yes. Yes. Yes. Yes. Yes. Yes. Yes.
Yes.
Yes. Yes. Yes. Yes. Yes. Yes. Yes.
Yes.
Yes. Yes. Yes. Yes
Yes.
Yes. Yes. Yes. Yes. Yes. Yes. Yes.
Yes.
Yes. Yes. Yes. Yes. Yes. Yes. Yes. 嗯。 Yes.
Yes.
Yes. Yes.
Yes.
Yes. Yes.
Yes.
Yes. Yes. 嗯.
Yes.
Yes. 嗯。
Yes.
Yes. 嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯.
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯。
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯。
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.
Yes.
嗯.