Quick Summary
Covering the latest research from top Wall Street investment banks

Rotation in China mega-cap cloud service providers still has room to run, with earnings and relative volatility offering the main trading opportunities

Institution
JPMorgan
Date
2026-08-10
Authors
Tony SK Lee;Haoshun Liu;Xipu Han;Twinkle Mehta, CFA;Davide Silvestrini;Bram Kaplan, CFA;Yangyang Hou
Company
-
Ticker
-
Industry
Internet platforms, information technology, and equity derivatives
Rating
-
NeutralLow confidencePolicy support, improving earnings expectations, and no obvious chase higher in options skew suggest positioning in China internet platforms remains uncrowded; meanwhile, realized volatility of single stocks relative to indices may still be underestimated.
AuthorsTony SK Lee;Haoshun Liu;Xipu Han;Twinkle Mehta, CFA;Davide Silvestrini;Bram Kaplan, CFA;Yangyang Hou
CoverageAsia-Pacific
Business segmentsChina internet mega-cap cloud service providers、Hong Kong technology stocks、Hong Kong structured products、Equity volatility strategies
Research firm divisions/subsidiariesJPMorgan(Other)、J.P. Morgan Securities (Asia Pacific) Limited(Other)、J.P. Morgan Broking (Hong Kong) Limited(Other)、J.P. Morgan Securities plc(Other)、J.P. Morgan Securities LLC(Other)

AI summary card

Rotation in China mega-cap cloud service providers still has room to run, with earnings and relative volatility offering the main trading opportunities

JPMorgan believes the rally in China internet platforms has not yet resulted in crowded positioning, recommends participating in upside through index and single-stock call spreads, going long volatility of mega-cap cloud service providers relative to HSCEI, and adopting a volatility monetization strategy for Lenovo.

The overall view is bullish on China internet mega-cap cloud service providers, with a preference for Tencent catch-up, and bullish on the volatility of Alibaba, Tencent, and Baidu relative to HSCEI; for Lenovo, the preference is to sell overvalued earnings volatility.
China mega-cap cloud service providersTencentAlibabaBaiduLenovoEarnings volatilityCall spreadsRelative volatilityHong Kong structured products
  • The three-month weighted average call option skew of major China internet stocks has changed little after the rally, indicating that upside positioning remains light.
  • The current market-cap-weighted average earnings-implied move in the Hong Kong market is about 2.3%, still below the historical average realized move of 2.7%.
  • Tencent still lags HSTECH year to date, with relatively solid fundamentals, and earnings may drive a catch-up move, but its short-term implied volatility is already at a historical high.
  • Lenovo's earnings-implied one-day move is about 8.9%, significantly above the three-year historical median of 3.2%, making it more suitable for selling volatility rather than directly chasing upside.
  • Hong Kong structured product issuance rebounded in July, and realized dispersion between single stocks and indices remains strong, supporting relative volatility trades.

Report interpretation

Overview

The report focuses on Hong Kong technology stocks' earnings season, the spread of China's artificial intelligence trade from hardware to internet mega-cap cloud service providers, and equity derivatives pricing. JPMorgan believes that support from the July Politburo meeting for the regulated and healthy development of internet platforms, improved earnings revision breadth in the technology and communication services sectors, and the absence of obvious chase higher in the options market jointly create conditions for further rotation into China internet platforms. In terms of implementation, the report prefers using call spreads on HSCEI, HSTECH, and JPCHINTE Index to obtain risk-defined upside exposure; for single stocks, it favors Tencent's post-earnings catch-up opportunity and recommends going long volatility of Alibaba, Tencent, and Baidu relative to HSCEI. Because Lenovo's earnings volatility pricing is clearly elevated, the report is more inclined to monetize volatility through selling put options or accumulator structures.

Core views

First, market leadership in China's artificial intelligence trade is spreading from hardware beneficiaries such as semiconductors to internet platforms with traffic, distribution, and monetization capabilities. Second, call option skew on China internet stocks has not steepened significantly after the rally, indicating that investor positioning is not crowded; if Alibaba's cloud business margin and overall earnings expectations inflect, fundamentals could still drive the next leg higher. Third, although Tencent has rebounded about 10% from its July 24 low, it remains without a significant year-to-date gain, and earnings could become a catch-up catalyst; given its implied volatility is at a high percentile, call spreads or knock-out calls are more efficient than directly buying vanilla calls. Fourth, Lenovo's share price has risen more than 100% since the last earnings release, but current earnings-implied volatility is far above its historical median, so investors should prioritize collecting option premium and setting disciplined dip-buying levels. Fifth, realized volatility and dispersion of Hong Kong single stocks relative to indices remain high, supporting medium-term relative volatility positions.

Analysis framework

The report combines policy and fundamental catalyst analysis, industry earnings revision breadth, relative performance of indices and single stocks, option skew, term structure, comparison of implied and realized earnings volatility, structured product issuance, southbound capital holdings, and regression and market regime classification. Trade recommendations further incorporate premium cost, Greeks, volatility percentiles, and historical earnings-day moves to select call spreads, knock-out calls, short puts, accumulator structures, and corridor variance swaps.

Methodology notes

  • Event volatility analysisComparison of earnings-implied volatility and historical realized volatility

    Compares the option-implied earnings-day price move with the median and range of actual earnings-day moves over the past three years.

    This method is used to identify the relative expensiveness or cheapness of earnings event volatility, but historical realized volatility does not guarantee future results; the report also notes that realized moves during the July to September earnings season may historically be higher than ex-ante implied pricing.

  • Positioning and sentiment analysisCall option skew analysis

    Assesses post-rally upside demand and crowding through the three-month call option skew of major China internet stocks.

    A rise in share prices without obvious re-leveraging of skew means the options market has not yet shown aggressive upside chasing, and fundamental improvement may still drive subsequent performance.

  • Relative value analysisSingle-stock versus index relative volatility

    Goes long volatility of Alibaba, Tencent, and Baidu while shorting HSCEI volatility.

    The report uses corridor variance swaps to express the view that realized volatility of single stocks is underestimated relative to the index, while controlling path risk through ranges, capped multiples, and sampling rules.

  • Statistical and regime analysisTurnover and forward earnings change regime model

    Uses stock turnover and next-twelve-month earnings changes to explain the realized volatility spread between mega-cap cloud service providers and HSCEI.

    The two variables have explained about 55% of changes in relative realized volatility since 2016; high turnover and improving earnings regimes have historically corresponded to higher and more persistent single-stock volatility relative to index volatility.

  • Structured trade designRisk-defined and volatility monetization strategies

    Selects call spreads, knock-out calls, short puts, or accumulator structures based on implied volatility levels.

    When implied volatility is high, premium costs can be reduced or option premium collected by capping part of the upside or taking on a purchase obligation at a preset price, but investors need to bear knock-out, tail downside, and path-dependency risks.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

  • HSCEI and HSTECH
    Serve as broad upside exposure instruments for China's large internet platforms and mega-cap cloud service providers
    Strengths
    Index implied volatility is moderate relative to historical ranges, and call spreads can limit premium costs.
    Weaknesses
    Diversified index constituents may dilute returns from earnings improvement at a single internet platform.
    Comparison
    Compared with JPCHINTE Index, they have broader coverage and lower concentration in a single theme.
    Risks
    Macro policy changes, a decline in overall risk appetite, and insufficient index upside may lead to loss of option premium.
  • JPCHINTE Index
    Used to express the theme of monetization and earnings improvement in China internet platforms in a more concentrated way
    Strengths
    Provides more direct exposure to mega-cap cloud service providers and the platform economy.
    Weaknesses
    Theme concentration is relatively high, and the relevant basket was created by J.P. Morgan's trading desk.
    Comparison
    Compared with HSCEI and HSTECH, it is more sensitive to fundamental inflection points in China internet companies.
    Risks
    Risks include constituent concentration, liquidity, pricing, and potential conflicts of interest from the product provider.
  • Tencent (700 HK)
    A target for catch-up in the China internet sector and an earnings catalyst
    Strengths
    Fundamentals are relatively solid, it remains a long-term sector top pick, and buybacks can resume after earnings and provide downside support.
    Weaknesses
    The earnings path may not be smooth, and two-month implied volatility is at a high percentile over the past three years.
    Comparison
    Year-to-date performance lags HSTECH, but the recent rebound from the low is clearly stronger than the index.
    Risks
    Earnings below expectations, insufficient catalysts, an extended buyback suspension period, and call spread or knock-out structures limiting returns.
  • Alibaba (9988 HK)
    A key stock to watch for earnings revisions among China mega-cap cloud service providers and improvement in cloud business margins
    Strengths
    Earnings forecasts have been raised for the first time since May 2025 and may enter a sustained positive revision cycle.
    Weaknesses
    Recent unit economics, price competition, and improvement in cloud business margins still need to be validated by earnings.
    Comparison
    Its fundamental inflection point can serve as an important leading signal for overall improvement among China mega-cap cloud service providers.
    Risks
    Cloud business margins missing expectations, insufficient progress in AI monetization, and intensifying competition.
  • Baidu (9888 HK)
    A long component of the China mega-cap cloud service providers versus HSCEI volatility basket
    Strengths
    If it completes conversion to dual-primary listing and is included in Stock Connect, turnover and southbound participation may increase.
    Weaknesses
    The report provides fewer company fundamental details than for Tencent and Alibaba.
    Comparison
    Its main value lies in single-stock realized volatility relative to the index, rather than purely directional upside.
    Risks
    Progress on Stock Connect inclusion, actual turnover, and earnings volatility falling below expectations.
  • Lenovo (992 HK)
    A target for confirming AI server fundamentals and monetizing earnings volatility
    Strengths
    AI server growth, improving server profitability, and continued southbound capital preference provide fundamental support.
    Weaknesses
    The share price has risen more than 100% since the last earnings release, and earnings-implied volatility is significantly above the historical median.
    Comparison
    Unlike Tencent's catch-up strategy, Lenovo is more suitable for selling overvalued volatility rather than directly buying upside options.
    Risks
    Selling put options or using accumulator structures may generate significant losses and forced delivery risk if the share price drops sharply.
  • Hong Kong single-stock volatility
    Relative value opportunity amid recovering structured product demand and stronger realized dispersion
    Strengths
    The realized volatility spread of the top fifteen HSCEI constituents relative to the index is close to the high since 2021.
    Weaknesses
    Increased structured product issuance will continue to supply single-stock volatility to the market.
    Comparison
    Single-stock implied volatility remains at historical highs relative to the index, but realized dispersion has also strengthened at the same time.
    Risks
    A sudden rise in index volatility, higher single-stock correlations, or a further increase in structured product supply.

Key data

  • Relative performance of China technology and semiconductorsSince late June, the China semiconductor index has pulled back about 20%, while HSTECH has risen 8.6%Shows that leadership in the artificial intelligence trade is spreading from hardware to internet mega-cap cloud service providers.
  • Hong Kong earnings peak periodFrom August 10, 2026, the busiest week will see about 30 companies report results, accounting for roughly 45% of the combined market capitalization of HSI and HSCEIEarnings results may become a catalyst for further market rotation.
  • Market pricing of earnings volatilityThe current market-cap-weighted average implied move is about 2.3%, while the historical average realized move is about 2.7%Although overall event risk pricing is higher than historical ex-ante implied levels, it remains below historical realized levels.
  • Tencent earnings volatilityImplied earnings-day move is about 3.2%, versus a three-year historical median of 1.0%Tencent event volatility is expensive relative to history.
  • Tencent two-month implied volatility33.2%, around the 86th percentile over the past three yearsThe cost efficiency of directly buying vanilla calls is relatively low.
  • Tencent recent relative performanceRebounded about 10% from the July 24, 2026 low, while HSCEI rose 2.5% and HSTECH rose 4.2% over the same periodIndicates that capital is rebuilding exposure to mega-cap cloud service providers, but this has not yet constituted sustained re-rating.
  • Tencent share buybackRepurchased about CNY15bn from late June to early July 2026Buybacks created stable demand in a weak market; after buybacks were suspended ahead of earnings, downside volatility rose again.
  • Tencent call spreadBuy 700 HK Oct26 105% to 125% call spread, quoted at 3.68%Delta is about 30.1%, with implied volatility at the two strikes about 33.6% and 36.5%.
  • Tencent knock-out call optionBuy 700 HK Oct26 105% call option with 125% knock-out barrier, quoted at 2.01%Saves about 45% premium versus a vanilla call spread, but upside is limited by the barrier feature.
  • Lenovo earnings volatilityImplied earnings-day move is about 8.9%, versus a three-year historical median of 3.2%The report identifies Lenovo as one of the stocks with the most expensive implied volatility relative to historical realized volatility.
  • Lenovo fundamentals and valuationRating upgraded from Neutral to Overweight, with a target price of HK$30, based on 16x next-twelve-month diluted EPSKey drivers are improving profitability in AI servers, better-than-feared IDG price elasticity, and industry consolidation amid component shortages.
  • Lenovo southbound capital holdingsOnce accounted for about 21% of free float, recently about 17%Although there has been profit-taking, the level remains significantly higher than the previous typical level of below 10%.
  • Explanatory power of the relative volatility modelStock turnover and next-twelve-month earnings changes explain about 55% of changes in the relative realized volatility spreadThe sample period starts from 2016.
  • Medium-term relative volatility rangeThe nine-month realized volatility spread between mega-cap cloud service providers and HSCEI has historically been 16 to 26 volatility pointsCorresponds to high turnover and improving earnings regimes, supporting medium-term relative volatility strategies.

Impact & implications

For directional investors, China internet platforms remain the main beneficiaries of the spread of the artificial intelligence theme from capital expenditure to monetization, but in a high event-volatility environment, call spreads should be prioritized to control premium spending. For volatility investors, index volatility is relatively moderate, while single-stock earnings, turnover, and earnings revisions may sustain higher realized volatility, making long-short volatility combinations of Alibaba, Tencent, and Baidu versus HSCEI more attractive. Lenovo shows the opposite characteristics: the fundamental outlook is positive, but event volatility is already expensive, making it suitable to sell volatility when investors can tolerate the risk of taking delivery. The recovery in Hong Kong structured product demand may increase the supply of single-stock volatility, but realized dispersion remains strong and has not yet eliminated relative value opportunities.

Risks

  • China internet platforms' earnings, cloud business margins, or progress in AI monetization may fall short of expectations, potentially interrupting the rotation trade.
  • Policy support may be weaker than expected, or the regulatory environment may tighten again.
  • The maximum payoff of call spreads is limited by the upper strike, so they may significantly underperform spot if the underlying rallies sharply.
  • Knock-out calls may terminate early if the underlying reaches the barrier, causing investors to lose subsequent upside participation.
  • Selling Lenovo put options or using accumulator structures may generate nonlinear losses and delivery obligations if the share price falls sharply.
  • Corridor variance swaps involve path dependency, range sampling, liquidity, valuation, and counterparty risks.
  • Historical relationships between implied and realized volatility and statistical regimes do not guarantee continued validity in future earnings windows.
  • Increased Hong Kong structured product issuance may depress single-stock implied volatility and weaken returns from long volatility strategies.
  • Buyback suspension, southbound capital profit-taking, or a decline in overall risk appetite may weaken downside support for single stocks.

What to watch

  • Whether Alibaba management can confirm sustained improvement in cloud business margins and drive earnings forecasts into a positive revision cycle.
  • Whether Tencent's earnings can become a clear catalyst for catch-up relative to HSTECH, and when share buybacks resume after earnings.
  • Whether Lenovo's AI server revenue, order momentum, pricing, and profitability are sufficient to validate the current re-rating.
  • Whether three-month call option skew on China internet stocks starts to steepen significantly, to assess whether positioning is becoming crowded.
  • Whether actual price moves during earnings season can continue to exceed moves implied by the options market.
  • Whether HSTECH's strength relative to China semiconductor stocks continues.
  • Hong Kong structured product issuance volume, the composition of put accumulator and accumulator products, and changes in single-stock volatility supply.
  • Southbound capital holding trends in Tencent, Lenovo, and other China internet platforms.
  • Progress on Baidu's dual-primary listing conversion and potential Stock Connect inclusion.
  • Whether the realized volatility spread between mega-cap cloud service providers and HSCEI can remain at a high-teens level.
Zhejiang ICP No. 2022035445-5
Disclaimer: Market data, charts, indicators, research views, and other information provided on this website are intended solely for information display, research communication, and educational reference. They should not be regarded as personalized investment advice, securities recommendations, trading instructions, solicitations, or guarantees of return. While we strive to improve the reliability of our data and content, such information may still be subject to delays, errors, incompleteness, or untimely updates due to source differences, methodological limitations, system processing, or market volatility. Users should exercise independent judgment based on their own circumstances and bear all risks and responsibilities arising from the use of this website.

Settings

Sign in to view recent logins