--- title: "Tencent Shifts Valuation Anchor with RMB 52.8 Billion Investment" type: "News" locale: "en" url: "https://longbridge.com/en/news/295674748.md" description: "Tencent's free cash flow turned negative due to massive capital expenditures, prompting the market to reassess its valuation framework. Despite significant AI investments and strong operating profit growth, spending is outpacing earnings, leading JPMorgan to lower its EPS Estimate for 2026. The article compares Tencent's current situation to Meta's high-spending period in 2022, suggesting that if AI can boost advertising revenue as it did for Meta, the stock still has upside potential" datetime: "2026-08-12T13:47:48.000Z" locales: - [zh-CN](https://longbridge.com/zh-CN/news/295674748.md) - [en](https://longbridge.com/en/news/295674748.md) - [zh-HK](https://longbridge.com/zh-HK/news/295674748.md) --- # Tencent Shifts Valuation Anchor with RMB 52.8 Billion Investment Over the past decade, Tencent's most solid persona was that of a "money-printing machine"—with net cash consistently in the hundreds of billions and quarterly share buybacks backed by real money. Now, this money-printing machine has started pouring out funds on a large scale, with the rate of spending exceeding the rate of earning. The market needs to answer a question that has never been seriously asked before: **What framework should be used to price Tencent's valuation?** Single-quarter operating cash flow of RMB 52.7 billion was offset by RMB 59.3 billion in capital expenditure payments, RMB 5 billion in media content payments, and RMB 2.2 billion in lease liability payments, pushing free cash flow into negative territory. Excluding prepayments for computing power, it remained positive at RMB 37.6 billion. Operations are fine, and cash is still being generated; the problem is the speed at which money is being spent. **JPMorgan has lowered its adjusted EPS Estimate for Tencent in 2026 by 5%. For full-year 2026 capital expenditures, Bank of America sees RMB 185 billion, while JPMorgan sees RMB 200 billion. This money will likely be burned continuously for three years.** In other words, **the initial returns on Tencent's AI investment are becoming visible, with both advertising and cloud segments using incremental revenue to pay down the debt**. Non-IFRS operating profit in 26Q2 was RMB 75.6 billion, a 9% year-on-year increase; if the impact of investments in new AI products (Hy, Yuanbao, WorkBuddy, CodeBuddy, Xiaowei) is excluded, Non-IFRS operating profit would have increased by 19% year-on-year to RMB 86.1 billion, with an operating margin of 42%. Tencent's current situation is not unfamiliar to US stock investors. In 2022, Meta increased its capital expenditures from approximately $19 billion to $32 billion (a year-on-year increase of about +68%). Free cash flow halved that year, and the stock price fell from $331 to $91 within 11 months, wiping out 70% of its market value. The market narrative at the time was identical to today's doubts about Tencent: "Metaverse money pit," "advertising slowdown," and "collapse of valuation logic." The subsequent script played out as follows: In 2023, Meta controlled costs through layoffs, its advertising business recovered, and AI recommendation models began to feed back into advertising revenue. The stock embarked on a five-fold rally over three years, rising from $91 to above $600. In the latest quarter, advertising revenue reached $55 billion, accounting for over 90% of total revenue, with net profit margins hitting record highs. Currently, Meta has raised its 2026 capital expenditure guidance to $125–145 billion, double that of 2025—the market did not panic, instead giving an 8.8% gain on the day Meta announced it would rent out idle GPU computing power. **Of course, Tencent is not a replica of Meta.** Meta relies on globalized US dollar advertising revenue, whereas Tencent's advertising is already highly monetized, with less incremental space for Video Accounts, Moments, and Mini Programs compared to Meta at that time; Tencent's international games also face exchange rate fluctuations and regulatory constraints simultaneously. But there is one common key variable: **The quarter when AI transitions from a "cost item" to a "revenue item" is the watershed moment for the shift in valuation narrative**. For Meta, it was 2023; Tencent appears to be on the eve of this critical point—with advertising up 22% and Cloud AI accounting for 25–30% of revenue, signs of realization have emerged in both directions. After the Hong Kong stock market closed on August 12, Tencent's second-quarter report looked good across almost every metric: Revenue was RMB 204.785 billion, an 11% year-on-year increase, returning to double-digit growth; Advertising was RMB 43.565 billion, up 22% year-on-year; Domestic Games were RMB 47.3 billion, up 17% year-on-year, with "Roco Kingdom: World" topping both the daily active users and gross billing charts for new games in the first half. Non-IFRS net profit attributable to shareholders was RMB 68.415 billion, a 9% year-on-year increase, in line with expectations. The only glaring issue was at the bottom of the cash flow statement: **Single-quarter capital expenditures were RMB 52.784 billion, a 176% year-on-year increase; free cash flow was -RMB 13.8 billion, turning negative for the first time in many years**. Total capital expenditures for the first half amounted to RMB 84.72 billion, an 82% year-on-year increase, already surpassing the full-year level of 2025 (RMB 79.2 billion). Over the past decade, Tencent's most solid persona was that of a "money-printing machine"—with net cash consistently in the hundreds of billions and quarterly share buybacks backed by real money. Now, this money-printing machine has started pouring out funds on a large scale, with the rate of spending exceeding the rate of earning. The market needs to answer a question that has never been seriously asked before: **What framework should be used to price Tencent's valuation?** Let's first look at where the money went. Of the RMB 52.784 billion in capital expenditures, the bulk was for AI computing power procurement—GPUs, data centers, and computing power prepayments. The company provided its own breakdown: Operating cash flow of RMB 52.7 billion was offset by RMB 59.3 billion in capital expenditure payments, RMB 5 billion in media content payments, and RMB 2.2 billion in lease liability payments, resulting in negative free cash flow; however, after excluding prepayments for AI-related computing power procurement, free cash flow was positive at RMB 37.6 billion. These figures clarify two things. First, the spending is proactive, not a sign of operational deterioration—if AI prepayments are excluded, cash flow remains healthy. Second, the intensity of investment has jumped: Single-quarter spending of RMB 52.8 billion is equivalent to two-thirds of the full-year 2025 amount (RMB 79.2 billion); institutional forecasts for full-year 2026 have been raised to RMB 185 billion (Bank of America) to RMB 200 billion (JPMorgan). This is not Tencent's first major investment, but there is a fundamental difference this time. Previously, it was "small steps, fast running": Game R&D, Video Account subsidies, and cloud infrastructure were all investments in the billions, reversible if needed. Now it is a "platform-level leap": Annual capital expenditures of RMB 200 billion imply that Tencent has accepted that AI is the core battlefield for the next three years, and that this money will need to be burned continuously. JPMorgan has already lowered its adjusted EPS Estimate for Tencent in 2026 by 5%, citing the raise in capital expenditures to RMB 200 billion, with depreciation and expenses rising in sync in the second half, potentially causing a significant contraction in free cash flow. The market's fear of "burning money" often stems from "seeing no return on the money burned." The most noteworthy aspect of Tencent's financial report is precisely that AI has begun to realize returns on the revenue side, though this is buried in segment data rather than listed as a separate item. Advertising business grew 22% year-on-year, the fastest among the three major segments. The core of this growth is the AI advertising system AIM+—covering WeChat Store, short dramas, mini-games, and other WeChat ecosystem advertising scenarios, with customer coverage exceeding 80%. AI creative capabilities have increased material production efficiency by nearly 80%. Management provided earlier figures during the earnings call: Advertising in 26Q1 grew 20% year-on-year, explicitly driven by AI recommendation models. Cloud business provides more direct evidence. Enterprise service revenue in 26Q1 grew 20% year-on-year, with cloud revenue growing 40% year-on-year—far higher than the group's overall growth rate; AI-related services now account for 25% to 30% of cloud revenue, and cloud gross margin has risen to over 35%. Fintech and Enterprise Services revenue in 26Q2 grew 9% year-on-year; while the growth rate remained unchanged, the structure shifted: Cloud, AI computing demand, and inference calls for WorkBuddy and CodeBuddy are all contributing new revenue. In other words, **the initial returns on Tencent's AI investment are becoming visible, with both advertising and cloud segments using incremental revenue to pay down the debt**. Non-IFRS operating profit in 26Q2 was RMB 75.6 billion, a 9% year-on-year increase; if the impact of investments in new AI products (Hy, Yuanbao, WorkBuddy, CodeBuddy, Xiaowei) is excluded, Non-IFRS operating profit would have increased by 19% year-on-year to RMB 86.1 billion, with an operating margin of 42%. Tencent's current situation is not unfamiliar to US stock investors. In 2022, Meta increased its capital expenditures from approximately $19 billion to $32 billion (a year-on-year increase of about +68%). Free cash flow halved that year, and the stock price fell from $331 to $91 within 11 months, wiping out 70% of its market value. The market narrative at the time was identical to today's doubts about Tencent: "Metaverse money pit," "advertising slowdown," and "collapse of valuation logic." The subsequent script played out as follows: In 2023, Meta controlled costs through layoffs, its advertising business recovered, and AI recommendation models began to feed back into advertising revenue. The stock embarked on a five-fold rally over three years, rising from $91 to above $600. In the latest quarter, advertising revenue reached $55 billion, accounting for over 90% of total revenue, with net profit margins hitting record highs. Currently, Meta has raised its 2026 capital expenditure guidance to $125–145 billion, double that of 2025—the market did not panic, instead giving an 8.8% gain on the day Meta announced it would rent out idle GPU computing power. Of course, Tencent is not a replica of Meta. Meta relies on globalized US dollar advertising revenue, whereas Tencent's advertising is already highly monetized, with less incremental space for Video Accounts, Moments, and Mini Programs compared to Meta at that time; Tencent's international games also face exchange rate fluctuations and regulatory constraints simultaneously. But there is one common key variable: **The quarter when AI transitions from a "cost item" to a "revenue item" is the watershed moment for the shift in valuation narrative**. For Meta, it was 2023; Tencent appears to be on the eve of this critical point—with advertising up 22% and Cloud AI accounting for 25–30% of revenue, signs of realization have emerged in both directions. Goldman Sachs made a statement before the earnings report: Tencent's expected P/E ratio for 2026 is around 12x, indicating that the valuation multiple has bottomed out. The subtext is that the market has hardly priced in Tencent's AI assets. Let's break it down. Tencent's forecast net profit for 2026 corresponds to a 12x P/E ratio, while the Hang Seng Tech Index's rolling P/E ratio is around 20–23x (less than 40% of its valuation percentile over the past 5 years); the average institutional target price for Tencent is around HKD 662, implying approximately 43% upside from the current stock price. This price gap is not due to market stupidity, but rather differences in pricing assumptions: Bears view the RMB 200 billion capital expenditure as a profit black hole, while bulls see it as a low-cost way to buy AI options. **The divergence will ultimately be resolved by data. Looking forward, judging whether Tencent's "capex for growth" strategy holds water requires monitoring just three quarterly signals.** The first is advertising growth rate. If Q3 advertising growth stays above 15% (indicating AI-driven organic growth even after excluding the 618 seasonal effect), it confirms the monetization logic of AIM+; if it falls below 12%, the contribution of AI comes into question, and seasonal explanations no longer hold. The second is the quantifiability of Cloud AI revenue. If management discloses Cloud AI revenue or similar metrics like ARR in the Q3 earnings call, the market will immediately reprice—Alibaba's MaaS ARR path of RMB 30 billion is already laid out, and Tencent Cloud AI's share of 25–30% of revenue is substantial; what is missing is only public disclosure. The third is the marginal change in free cash flow. If the quarterly level of RMB 37.6 billion (excluding AI prepayments) can be maintained, it indicates that cash flow is being "actively surrendered" rather than suffering from "continuous bleeding"; if it turns negative even after exclusion, coupled with net cash continuing to decline from RMB 58.2 billion, the "cash cow" narrative will truly become unsustainable. For holders, the greatest significance of this financial report is the declaration of a new rule of the game: Henceforth, Tencent's valuation will no longer be determined solely by "profit and free cash flow," but by "AI revenue visibility." Buying at 12x PE is buying uncertainty; once Q3 data is released, the market will choose sides again between 12x and 20x. ### Related Stocks - [TCEHY.US](https://longbridge.com/en/quote/TCEHY.US.md) - [TCTZF.US](https://longbridge.com/en/quote/TCTZF.US.md) - [00700.HK](https://longbridge.com/en/quote/00700.HK.md) - [JPM.US](https://longbridge.com/en/quote/JPM.US.md) - [BAC.US](https://longbridge.com/en/quote/BAC.US.md) - [META.US](https://longbridge.com/en/quote/META.US.md) - [09988.HK](https://longbridge.com/en/quote/09988.HK.md) - [BABA.US](https://longbridge.com/en/quote/BABA.US.md) - [GS.US](https://longbridge.com/en/quote/GS.US.md) - [STECH.HK](https://longbridge.com/en/quote/STECH.HK.md) - [80700.HK](https://longbridge.com/en/quote/80700.HK.md) - [HTCD.SG](https://longbridge.com/en/quote/HTCD.SG.md) - [JPM-M.US](https://longbridge.com/en/quote/JPM-M.US.md) - [JPM-C.US](https://longbridge.com/en/quote/JPM-C.US.md) - [JPM-D.US](https://longbridge.com/en/quote/JPM-D.US.md) - [JPM-L.US](https://longbridge.com/en/quote/JPM-L.US.md) - [8634.JP](https://longbridge.com/en/quote/8634.JP.md) - [JPM-K.US](https://longbridge.com/en/quote/JPM-K.US.md) - [JPM-J.US](https://longbridge.com/en/quote/JPM-J.US.md) - [BAC-Q.US](https://longbridge.com/en/quote/BAC-Q.US.md) - [BML-L.US](https://longbridge.com/en/quote/BML-L.US.md) - [BAC-L.US](https://longbridge.com/en/quote/BAC-L.US.md) - [BAC-K.US](https://longbridge.com/en/quote/BAC-K.US.md) - [BML-H.US](https://longbridge.com/en/quote/BML-H.US.md) - [BAC-N.US](https://longbridge.com/en/quote/BAC-N.US.md) - [BAC-E.US](https://longbridge.com/en/quote/BAC-E.US.md) - [MER-K.US](https://longbridge.com/en/quote/MER-K.US.md) - [BAC-O.US](https://longbridge.com/en/quote/BAC-O.US.md) - [BML-G.US](https://longbridge.com/en/quote/BML-G.US.md) - [BAC-M.US](https://longbridge.com/en/quote/BAC-M.US.md) - [BAC-B.US](https://longbridge.com/en/quote/BAC-B.US.md) - [BML-J.US](https://longbridge.com/en/quote/BML-J.US.md) - [BAC-S.US](https://longbridge.com/en/quote/BAC-S.US.md) - [BAC-P.US](https://longbridge.com/en/quote/BAC-P.US.md) - [8648.JP](https://longbridge.com/en/quote/8648.JP.md) - [89988.HK](https://longbridge.com/en/quote/89988.HK.md) - [HBBD.SG](https://longbridge.com/en/quote/HBBD.SG.md) - [W4VR.SG](https://longbridge.com/en/quote/W4VR.SG.md) ## Related News & Research - [Another AI lab is burning through cash as Tencent earnings rise](https://longbridge.com/en/news/295641594.md) - [AI startup Manus to resume independent operations as deal with Meta unwinds](https://longbridge.com/en/news/295555892.md) - [What's Going On with the Drop in Meta Stock?](https://longbridge.com/en/news/295700719.md) - [Meta Platforms (META) Stock Dropped, So What Is Driving Attention Now?](https://longbridge.com/en/news/295721372.md) - [Four takeaways from Mark Zuckerberg’s massive AI manifesto](https://longbridge.com/en/news/295426243.md)