Thank you, Desmond. Hello, everyone, and welcome to Qifu Technology's third quarter 2023 earnings conference call. Our earnings release was distributed earlier today and is available on our IR website. Joining me today are Mr. Wu Haisheng, our CEO; Mr. Alex Xu, our CFO; and Mr. Zheng Yan, our CIO. Before we start, I would like to refer you to our safe harbor statement in the earnings press release, which applies to this call as we will make certain forward-looking statements. Also, this call includes discussions of certain non-GAAP financial measures. Please refer to our earnings release, which contains a reconciliation of non-GAAP financial measures to GAAP financial measures. Also, please note that unless otherwise stated, all figures mentioned in this call are in RMB terms. Before we start, please also note that today's prepared remarks from our CEO will be delivered in English using an AI-generated voice. Now I will turn the call over to Mr. Wu Haisheng. Please go ahead.

Hello, everyone. Thank you for joining our third quarter earnings conference call. Since the start of the year, the momentum of China's macroeconomic recovery has softened after rebounding earlier in the year, with effective demand for consumer credit also coming in weaker than expected. We promptly adjusted our strategic approach to this new market environment by diversifying our customer acquisition channels and refining operations, all with the clear goal of driving high-quality growth and improving profitability. These efforts yielded solid results in Q3. By the end of the quarter, our platform empowered a total of 155 financial institutions and cumulatively served more than 49 million users with approved credit lines. Total loan facilitation and origination volume on our platform reached RMB123.1 billion, up roughly 11% year-over-year. With the quality of our earnings further improving, our non-GAAP net income for the quarter increased by approximately 14% year-over-year. In today's complex macro environment, it is particularly important that we expand our safety cushion by improving profitability. To achieve this, we are optimizing resource allocation across customer acquisition, products, risk management, and asset distribution to boost operational efficiency and ultimately drive bottom-line growth. Now I'll walk you through some of the progress we made in this regard during the quarter. To start with, we continue to explore diversified customer acquisition channels and deploy innovative approaches to attract new customers, which not only improved customer acquisition efficiency but also resulted in notably better quality of new users. In terms of optimizing our customer acquisition channels, we established a partnership with a leading short-form video platform through our embedded finance business in the third quarter, leveraging the platform's massive active user base and our ability to accurately profile users and identify risk. We have consistently maintained a leading market share on the platform since the start. During the quarter, our embedded finance business generated an impressive 46% sequential increase in the number of new users with approved credit lines. Approximately 33% of total new users with approved credit lines during the quarter were acquired through the embedded finance channel. We also intensified our marketing efforts through innovative marketing approaches during the quarter to attract high-quality customers. This was done by deploying more compelling ad placements to expand our customer reach. Based on user profile analysis, users acquired through innovative marketing methods significantly outperformed those obtained through conventional methods in terms of educational background, mortgage, and credit history. They also clearly have better risk performance in terms of short-term delinquency rates as we observed during the first three months with our innovative customer acquisition strategy firmly in place. The proportion of high-quality users increased by more than 4 percentage points from July to October. This not only directly complements our existing user base but also contributes to the stability of our overall risk performance. As a result of these initiatives, the number of new users with approved credit lines increased by 18% in Q3 sequentially, while unit customer acquisition costs increased only slightly. Moving on to product design, we introduced a loyalty program to enhance engagement of existing users. By offering a wider range of value-added services, we are effectively enhancing user engagement and retention, which increases our revenue and profit per user. Users who joined the loyalty program demonstrated higher engagement with a double-digit increase in both the rate of drawdown and the number of loans borrowed over a certain period of time. Turning to risk, we swiftly responded to the changing market environment. In Q3, we gradually tightened credit standards and upgraded risk models. In particular, we further enhanced the knowledge graph for our broadly defined SME segment. By closely analyzing relationships among individuals, businesses, and industries, we improved our ability to identify risks within this segment. It's worth noting that there are clear differences between our broadly defined SME segment and SMEs in the traditional sense as primarily younger enterprises acquired through online channels. When conducting risk assessments, we place more emphasis on individual credit history and use business data as a secondary reference. As a result, credit lines extended to our SME segment are usually smaller, making associated risk much more manageable. Compared to the consumer segment, the SME segment generates much more stable demand and stronger growth potential with similar risk performances. Looking ahead, we are confident that there is significant room for growth and value generation over the long term for our broadly defined SME segment, driven by our accurate user identification and differentiated operations. Lastly, we kept pricing at stable levels and leveraged our funding strengths to further optimize our economic model. With relatively ample liquidity in the financial system during the quarter, we further optimized our funding structure by increasing the proportion from larger banks, which reduced our funding cost by another 20 basis points. In terms of ABS, we secured ABS investments from multiple state-owned and joint stock banks and top securities firms leading to a 47 basis point decrease quarter-over-quarter in ABS issuance costs. With the accuracy of user profiling and identification continuously improving, we also further expanded the range of our financial institution partners, strengthening our ability to serve various loan asset segments with more financial partners coming onboard under the ICE or referral model, which further mitigates risk. Our ability to engage with the lower-tier user segment has significantly improved. As we continue to optimize asset allocation efficiency, we expect our overall profitability to further improve going forward. Now let me share with you the progress we made on the technology front. Our technology solutions business is making solid progress, as we expand the array of solutions we offer to cover the entire credit process. During the quarter, we entered into partnerships with three additional financial institutions, each from a different category: a joint stock, internet, and private bank. To cater to the diverse needs of our banking partners, we adopted various deployment models and are committed to providing them with end-to-end technology solutions as well. Going forward, we will extend our end-to-end technology solutions to more financial institution partners to scale up our client base. In the long run, we expect a steady increase in the take rate for our technology solutions business. Additionally, we continued to invest in cutting-edge technologies such as artificial intelligence and large language models. On September 18, we partnered with the China Academy of Information and Communications Technology to officially unveil the first domestic standard for large language models in the financial sector. The standard will serve as a crucial reference and benchmark for the design, development, application, and review of large language models in finance. At the same time, we purchased hundreds of graphics cards and have been exploring the application of AI GC technology to improve efficiency throughout the entire credit process by conducting tests or implementing it in advertising, telemarketing, loan collection, and quality control. For example, our telemarketing system is now able to conduct semantic analysis and extract valuable leads from each conversation, improving our telemarketing conversion rate by more than 5%. In addition, around 70% of our image-based marketing materials are currently generated by AIGC. In the future, we will use large language models to tag and rate these materials from multiple dimensions to optimize ad placements and boost marketing effectiveness. Looking back at the past three quarters, despite a challenging macro environment, we remain vigilant about market conditions and improved our earnings quality by further optimizing operations and fine-tuning our business model. Since Q3, certain macro indicators are showing marginal improvement as a result of multiple supporting policies. Against this backdrop, we are confident in our ability to continuously make breakthroughs and create value for our shareholders with better results. With that, I will now turn the call over to our CFO, Alex Xu, who will walk you through our financial results for the quarter.

Okay. Thank you, Haisheng. Good morning, and good evening, everyone. Welcome to our third quarter earnings call. Facing slower-than-expected macro recovery and worsening consumer sentiment, we proactively took actions in Q3 to optimize our product and service offerings, strengthened relationships with users and key partners, aiming to drive long-term sustainable quality growth. In Q3, we saw some volatility in asset quality and key leading risk metrics start to fluctuate from historical best levels achieved in previous quarters. Day 1 delinquency was 4.6% in Q3 versus 4.2% in Q2. The uptick in day 1 delinquency mainly reflects borrowers' negative sentiment toward the ongoing macro uncertainties. The 30-day collection rate was 86.7% in Q3 versus 87.2% in Q2. This modest decline was also driven by macro weakness. Throughout Q3, we have proactively adjusted our risk management models and gradually applied more restrictive standards on incoming applications. By late September, we started to see stable credit quality among new borrowers. As economic conditions remain challenging, we may continue to see some fluctuation of these metrics in the near future, although overall risk levels should still be manageable with our continued effort to proactively mitigate risks. Total net revenue for Q3 was RMB 4.3 billion versus RMB 3.9 billion in Q2 and RMB 4.1 billion a year ago. Revenue from credit-driven service, capital heavy, was RMB 3.1 billion in Q3 compared to RMB 2.8 billion in Q2 and RMB 2.9 billion a year ago. The year-on-year growth was mainly due to growth in on-balance sheet loans partially offset by declines in expected average tenor of the loans. The sequential increase reflected growth in loan balance as well as continued improvement in effective tenors. On-balance sheet loans account for over 19% of total loan volume. Overall funding costs further declined by roughly 20 bps with the help of our strong relationship with financial institution partners as well as additional issuance of ABS. Revenue from platform service capital light was RMB 1.2 billion in Q3 compared to RMB 1.1 billion in Q2 and RMB 1.2 billion a year ago. The sequential growth was mainly due to continued improvement in overall effective tenor of the loans and strong contribution from ICE, substantially offsetting the decline in capital-light loan facilitation volume. For Q3, capital-light loan facilitation, ICE, and other tech solutions combined account for roughly 56% of the total loan volume compared to roughly 58% in the prior quarter. We expect this ratio to be roughly stable around this level for the year. We will continue to evaluate different components of our operation and seek a better mix between risk-bearing and non-risk-bearing solutions based on macro environment and operational conditions. In Q3, we saw continued sequential improvement in revenue take rate for both cap-heavy and cap-light business as early repayment ratio gradually return to normal levels and effective tenors gradually extended. During the quarter, average IRR of the loans we originated and/or facilitated remained stable quarter-on-quarter, well within the regulatory rate cap requirement. Looking forward, we expect pricing to be fluctuating in a narrow band around this level for the coming quarters. Sales and marketing expenses increased 21% quarter-on-quarter but declined 15% year-on-year. We added over 1.7 million new credit line users in Q3 compared to 1.5 million in Q2. The unit cost to acquire a new credit line user also increased modestly quarter-on-quarter to 306 from 296 in Q2. While we will continue to drive for efficiency in our operation, we may adjust the pace of our new user acquisition based on macro conditions from time to time. As Haisheng mentioned, we have made noticeable progress in diversifying our user acquisition channels during the quarter. Meanwhile, we will continue to focus on re-energizing existing user base as repeat borrowers historically contribute the vast majority of our business. As macro uncertainties persist and credit quality fluctuates, we will continue to take a prudent approach to book provisions against potential credit losses. Total new provision for risk-bearing loans in Q3 was approximately 2.1 billion, and write-backs of previous provisions were approximately 600 million. The provision coverage ratio, which is defined as total outstanding provision divided by total outstanding delinquent loan balance between 90 and 180 days was 534% in Q3 compared to 511% in Q2. Non-GAAP net profit was RMB 1.18 billion in Q3 compared to RMB 1.15 billion in Q2. The effective tax rate for Q3 was over 22% compared to our typical ETR of approximately 15%. The higher ETR in Q3 was mainly due to additional withholding tax provision related to cash distribution from onshore to offshore for dividend payments and share repurchase program. Please also note, our second quarter earnings were helped by a tax rebate of approximately RMB 160 million. Excluding the tax rebate, non-GAAP net income actually grew approximately 16% sequentially in Q3. With solid operating results and stable contribution from cap-light models, our leverage ratio, which is defined as risk-bearing loan balance divided by shareholders' equity, was 3.5 times in Q3, near historical low compared to 3.8 times a year ago. We expect to see the leverage ratio fluctuate around this level in the near future. We generated approximately RMB 1.2 billion cash from operations in Q3 compared to RMB 1.8 billion in Q2. The decline was mainly due to the change in working capital at the end of the quarter. Total cash and cash equivalents was RMB 8.2 billion in Q3 compared to RMB 8.5 billion in Q2. Non-restricted cash was approximately RMB 4.9 billion in Q3 compared to RMB 5.3 billion in Q2. The sequential decline in cash position was mainly due to increased cash usage in our on-balance sheet lending. As we discussed earlier, we will continue to look for opportunities to deploy resources to launch new initiatives, develop new technologies, and expand service offerings. As we continue to generate healthy cash flow from operations, we believe our current cash position is sufficient to support our business development and to return to our shareholders. In June 20, 2023, we announced a share buyback program to repurchase up to 150 million over a 12-month period. As of November 16, 2023, we have bought approximately $80 million worth of ADS in the open market at an average price around $16.2. We will continue to execute the buyback program in accordance with related rules and regulations. With the full execution of the repurchase program and the dividend plan, the combined payout ratio will exceed 50%. Going forward, we will continue to optimize our capital allocation plan and make timely adjustments to generate attractive returns to our shareholders. Finally, regarding our outlook. While macro recovery appears slower than expected, we remain confident to achieve our operational targets for the year. As such, we now expect Q4 total loan volume to be between RMB 116 billion and RMB 126 billion and for the full year, total loan volume to be between RMB 473 billion and RMB 483 billion, representing year-on-year growth of 15% to 17%. As always, this forecast reflects the company's current and preliminary view, which is subject to material changes. With that, I would like to conclude our prepared remarks. Operator, we can now take some questions.

For those who can speak Chinese, please ask your question in Chinese first, followed by an English translation. Additionally, to ensure we have enough time for everyone on the call, please limit your questions to one initial question and one follow-up. If you have more questions, please return to the queue. The first question comes from Richard Xu from Morgan Stanley. Please go ahead.

Essentially I have two questions. One is, exactly what's the credit demand trend at the moment. Particularly, there's some demand for consumption from the SME client base. What are the differences between the trends from the typical consumption loan demand and also any differences in strategy and risk management? Secondly, what's the funding cost and whether there's further room to reduce funding costs in the future? Thanks.

Okay. Thank you, Richard. I want to highlight that the prepared remarks just now delivered were generated by our large language model team. But for the Q&A part, I will still answer the question in Chinese. Regarding your question, stepping into Q4, we have observed the credit demand slightly trending down. In particular, credit demand was soft in October, mainly due to the seasonality factor during the national holiday, and after that, it maintained stable into November. From a user segment perspective, the credit demand from the broadly defined SME segment is slightly better than our consumer sector, and we do see some divergence from an industry perspective. For example, the service sector, like sports, entertainment, and the technology sector are better than the others. Compared to consumer segments, the broadly defined SME segment has more stable credit demand and generates higher value. We expect to enhance our ability to identify the user group and use diversified products and services to improve our customers' experience and their stickiness as well. Here, we would like to emphasize our broadly defined SME users are totally different from traditional SMEs as their ticket size is much smaller and our risk model is more based on individual profiles with the business data and supplementary information and therefore, the risk performance of our broadly defined SME users is much better. And I would like to invite our CIO, Mr. Zheng Yan, to add on the risk part.

As Haisheng just mentioned, our risk management mechanism for the broadly defined SME segment is different from the traditional SME sector, in the sense that our risk model is primarily based on personal data of the business owner with the enterprise data as the secondary layer of information input. Upon users' approval, we integrated personal credit data with the SME specific data sources, for example, the tax, invoice, cash flow, billings, etc., based on our sophisticated algorithm, so we can better evaluate the creditworthiness of the SME borrowers. As this segment normally has urgent and higher credit amount, we combined the fixed credit line with additional temporary credit lines in relation to certain seasonal events like the Double 11 Shopping Festival on June 16, etc., so we can better serve all kinds of needs of the SME users.

Regarding the funding cost, our cost of funding declined by 20 bps sequentially in Q3 with our ABS issuance costs down by 47 bps. Our current funding cost is quite similar to the Q3 level. From a rate-cutting perspective, theoretically, rate cutting can drive down the funding cost of financial institutions and eventually benefit the real economy. However, due to the complex macro environment this year, we didn't see it help that much. Our funding costs decline this year was mainly driven by the supply and demand dynamics in the market as the funding partners allocated more resources to consumer loan assets when the mortgage loan is underperforming this year. Given the scarcity of our assets, we managed to continue to optimize our funding costs this year. However, since September, there are some other factors, including government financing and property sectors are also attracting funding flows, which may put pressure on further reduction of our funding cost. We will continue to optimize our funding structure and keep the pace or even expand our ABS issuance to maintain our competitive advantage from the funding part and maintain our funding cost at a stable level.

I'll now move on to the next question from Alex Ye from UBS. Please go ahead.

So my first question is about the asset quality outlook. Can you give us more color in terms of the reason for causing fluctuations in Q3? Is it more driven by the macro environment or some of the industry-specific reasons such as your collection efficiency? And what's the latest trend in Q4 in the most recent months? And the second question is on the loan pricing outlook. So in September, we have seen China lowering the mortgage rate for the existing mortgages. So in that kind of a backdrop, do you expect the current lower interest rate environment to also translate into some of the downward pressure to our loan pricing in the future? Thank you.

Okay. I will do the translation. The fluctuation of our asset quality in Q3 was mainly a result of two factors adding together. The first is, from a macro perspective, the key macro indicators are below the expectation. The second reason is that the seasonality of the credit industry in the sense that liquidity is typically better in the first half than the second half, which led to some fluctuation of our asset quality in Q3. However, we expect the situation, especially the liquidity situation, will improve in early next year, which may help improve the risk performance of our new loans issued in Q4. At the same time, we swiftly adjusted our operations by upgrading our user base, tightening our credit assessment criteria and reducing the exposure of our existing borrowers since Q3. Accordingly, our early risk indicators for new loans say our FPD7 improved by 5% sequentially in September, and we expect it to continue to improve in October. In Q4, we have further strengthened our risk management tools from three aspects. First, we incorporated the scorecard data from three leading Internet platforms and built joint models with those platforms. Second, we deep-dived into our internal data and upgraded our regular B-scorecard by further leveraging our app data and PBOC data, using sequence algorithms and large models to construct risk sub-analysis aiming for an integrated B scorecard and further improved screening capability by the end of November. Third, we will upgrade our post-lending risk models and generate a fast responding B scorecard rating system by the end of December, primarily based on recent samples, which is more helpful for us to identify new risks. Putting together all these efforts, we expect the risk performance of the new loans originated in November and December will stabilize.

Regarding pricing, I would say, from our own intention perspective, we will conduct different pricing tests for various customer groups to improve user stickiness and satisfaction. Through more attractive pricing, we are also able to acquire quality new users to improve our user mix. Meanwhile, we will also raise our pricing for higher-risk users. Combined, our average pricing will continue to keep stable. Under the current macro environment, consumer finance is also one of the means to boost consumption and the economy. Under the current regulatory system, this is a multilayer supply system for the consumer credit industry. And based on that, we will also set price based on supply and demand dynamics and actively play our role to drive consumption and help SME players.

Next question comes from the line of Emma Xu of Bank of America Securities. Please go ahead.

So my question is about your shareholder capital return. It's really encouraging to see that you continue to buy back your shares at the current level. And I just want to ask whether you will continue to do the buyback. And after the completion of the current buyback program, will you launch more buyback programs in the future?

Okay. Emma, I will take this question. So as you said, we already did $80 million repurchase since June 20, well ahead of the timeline. And obviously, we will continue to execute the current repurchase program throughout the remainder of the year and also into next year. We already structured internally a very comprehensive system to periodically review our cash position as well as the efficient usage of the cash, trying to compare the expected return between the reinvestment in operations and the return generated from either the buyback or the dividend payment. Through this kind of review, we will determine which is the best way to deploy additional cash or additional resources in the future. So we, after we complete the current buyback program, as well as this year's dividend plan, we will do a new review based on the cash position at that point and make necessary changes to the shareholder return program either through a buyback or through a dividend or through some kind of a combination of the both. In summary, basically, the logic behind our future cash deployment is based on whichever method can generate the highest shareholder return. Then, we will use the cash in that direction.

Next question, we have the line from Cindy Wang from China Renaissance. Please go ahead.

Thanks management team for giving this opportunity. So my question is about the management strategy for capital-light and capital-heavy assets. Can you help us breakdown the current profit margins of capital-light and capital-heavy loans? And how does the management view the proportion of capital light and capital-heavy loans in the future? So compared with the growth of loan volume, will the company be more focused on profitability in 2024?

Okay. Thanks, Cindy. Regarding the profitability of our capital-heavy, capital-light business model, I would say both capital-heavy and capital-light loan facilitation model generates a roughly 3% net take rate with capital-heavy a little bit higher than 3% and capital-light a little bit lower but pretty close to 3%. In terms of the loan mix, we will balance profitability and the long-term healthiness of our loan portfolio and keep a dynamic balance between these two categories. Regarding the importance of loan volume versus profitability, I would say it is not our priority to either pursue loan growth or pursue profit improvement. If we only pursue profitability, we can just all do as a heavy business like some other platforms. But our target is to balance profitability and the long-term healthiness and sustainability of our loan portfolio. So we expect the loan mix to maintain largely stable in the foreseeable future.

Yes. I probably want to add one little point here, that this year, in 2023, because of the macro factors and also because the first half of the year, we were facing a tough competition in terms of pricing versus last year. That's why you see the profitability growth is slower than loan volume growth for 2023. And I would say, these negative factors related to profitability are already behind us. So in the future years, we will try to at least maintain the same pace in terms of between the volume and the profitability. If anything, we try to also drive some additional operating leverage in the future. Thank you, Cindy.

Our next question comes from the line of Yada Li from CICC. Please go ahead.

Then, I'll do my translation. Hello management, thanks for taking my question. This is Yada.

Okay. Because we understand your question, so we can go ahead and answer this question. Okay. Regarding the take rate, I would say, in the short run, we will see some fluctuation in terms of our credit costs and funding costs. In this regard, we will further fine-tune our operations, optimize our funding structure and invest more resources into R&D to improve our operational efficiency and funding costs. With those costs and efficiency improved, we will mitigate largely the fluctuation. So we will see the take rate maintain stable in the short term. Actually, if you look at our take rate in Q3, it's actually improved sequentially from an apple-to-apple perspective with diversified services offered to our users and optimizing our asset allocation. But because we expanded our customer outreach in Q3 and added on additional tax costs related to our dividend and buyback program, it looks like our take rate in Q3 maintained stable from last quarter. From a long-term perspective, we will further improve our operational efficiency, and we do hope to see a steady increase in our take rate.

I see no more questions from the phone line. I would like to hand the call back to management for closing remarks.

Okay. Thanks again for everyone joining us for the call. If you have additional questions, please feel free to contact us offline. Thank you. Have a good day.

Ladies and gentlemen, that concludes today's conference call. Thank you for your participation. You may now disconnect your lines.