A-share sentiment cools, while Hong Kong stocks are still seen as a more attractive recovery window

Overview

This report focuses on sentiment and allocation windows in the China equity market, with the core indicator being Morgan Stanley's A-share Sentiment Index (MSASI). The report shows that A-share investor sentiment weakened ahead of the July Politburo meeting and the CXMT IPO, mainly due to a policy vacuum, IPO-driven liquidity diversion, weaker-than-expected second-quarter GDP, and soft domestic demand and real estate data. At the same time, the report believes Hong Kong stocks have a more favorable near-term market structure, as the drag from e-commerce price competition may be nearing its end, AI commercialization continues to advance, volatility following certain share lock-up expiries has eased, and late July to August may still be an important window for a more sustainable recovery in Chinese equities.

Core views

First, A-share sentiment has cooled notably versus the previous period, with weighted MSASI falling to 37% and the one-month moving average falling to 57%, alongside declines in trading activity and margin balances. Second, the focus of policy expectations is not on new large-scale stimulus, but on accelerating the implementation of existing fiscal quotas while continuing to prioritize support for technological self-reliance, advanced manufacturing, and energy security. Third, near-term market dynamics in Hong Kong stocks are relatively more favorable than in A-shares, making them the main direction for increasing China equity exposure. Fourth, the long-term case for A-share technology and innovation opportunities has not been invalidated; if earnings growth continues to accelerate and global volatility eases, related sectors are expected to recover and break to new highs. Fifth, a more durable market recovery requires simultaneous improvement in several conditions, including earnings revisions, AI capital expenditure, the global interest-rate path, and deleveraging pressure.

Analysis framework

The report tracks A-share investor sentiment through the MSASI framework, standardizing 12 indicators including turnover, margin financing, futures, northbound trading, account openings, RSI, number of limit-up stocks, futures premium/discount, put-call ratio, passive fund flows, and earnings revisions, and weighting them by their historical explanatory power relative to the CSI 300. It then combines southbound flows, policy expectations, real estate data, technology earnings pre-announcements, and Hong Kong market catalysts to form its view on the short- to medium-term risk-reward in Chinese equities.

Methodology notes

• Market sentiment indicatorMSASI

  Morgan Stanley A-share Sentiment Index

  MSASI is built on 12 market sentiment and trading activity indicators to measure the strength of A-share investor sentiment. A higher reading indicates stronger investment enthusiasm or risk appetite, while a lower reading indicates weaker sentiment or a more pronounced risk-off bias.

• Standardization method100-day rolling min-max normalization

  Normalized Value=(Latest Value-Min Last 100 Days)/(Max Last 100 Days-Min Last 100 Days)

  Each indicator is first rescaled to a 0-100 range based on the past 100-day rolling window, making data with different units and frequencies comparable while highlighting changes in the medium-term sentiment direction.

• Weighting methodSingle-factor regression R-squared weights

  Weighted by each indicator's historical explanatory power relative to the CSI 300

  Each standardized indicator is weighted according to the R-squared from its single-factor regression against the CSI 300's performance relative to its 100-day moving average; indicators with stronger historical correlation receive higher weights in the composite sentiment index.

• Trend smoothingMSASI Weighted 1MMA

  One-month moving average of weighted MSASI

  The report applies a one-month moving average to weighted MSASI to reduce high-frequency volatility and more clearly reflect the medium-term sentiment trend.

Asset mapping & comparison

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

• A-shares

  Core research target

  Strengths

  Long-term opportunities in technology innovation and advanced manufacturing remain favored, and if earnings growth continues to accelerate, related sectors have the potential to recover and reach new highs.

  Weaknesses

  Short-term sentiment has cooled, with trading and margin activity declining, while facing uncertainty from a policy vacuum, liquidity diversion from the CXMT IPO, and global AI volatility.

  Comparison

  The report believes recent market dynamics in Hong Kong are more favorable than in A-shares.

  Risks

  Policy response weaker than expected, larger-than-expected liquidity drain from IPOs, continued negative earnings revisions, and ongoing drag from real estate and domestic demand.

• Hong Kong stocks

  More preferred allocation direction

  Strengths

  The report remains constructive on Hong Kong stocks, viewing current levels as an attractive entry point to increase exposure, with southbound funds maintaining net inflows and many technology and AI-related catalysts.

  Weaknesses

  Still affected by global volatility, the interest-rate path, and China macro expectations.

  Comparison

  Compared with A-shares, near-term market dynamics in Hong Kong stocks are seen as more favorable, mainly due to AI commercialization, easing e-commerce competition, and digestion of lock-up expiry volatility.

  Risks

  Repeated global deleveraging pressure, an unclear Fed rate-cut path, and weaker-than-expected AI commercialization.

• CSI 300

  MSASI weighting and comparison benchmark

  Strengths

  As a core broad-based A-share indicator, it is used to measure the correlation between sentiment indicators and market performance.

  Weaknesses

  Heavily influenced by domestic demand, real estate, earnings revisions, and policy timing.

  Comparison

  All MSASI indicators are assessed for explanatory power relative to the CSI 300 and its 100-day moving average.

  Risks

  If earnings forecast revisions fail to stabilize, the sustainability of index recovery may be insufficient.

• China technology and AI-related stocks

  Medium- to long-term recovery theme

  Strengths

  The report mentions positive updates such as Tencent's Hunyuan AI agent, Alibaba Cloud business, and EBITDA outlook, and believes AI commercialization and technological self-reliance remain important policy directions.

  Weaknesses

  In the short term, they are affected by global volatility, uncertainty in AI-related sectors, and disruptions from share lock-up expiries.

  Comparison

  Hong Kong technology catalysts are viewed as clearer in the near term than related A-share directions.

  Risks

  Insufficient evidence of AI capital spending, slower-than-expected commercialization progress, and spillover from global tech-stock volatility.

Key data

• Weighted MSASI37%Down 7 percentage points from the previous cutoff date of July 15, indicating weaker A-share sentiment.
• Weighted MSASI 1MMA57%Down 2 percentage points over the same period, indicating that the medium-term sentiment trend has also cooled.
• ChiNext turnoverRmb667bnDown 6% from the previous period.
• A-share turnoverRmb2,677bnDown 7% from the previous period.
• Margin financing and securities lending balanceRmb2,729bnDown 6% from the previous period.
• Stock index futures turnoverRmb734bnUp 8% from the previous period, in contrast to weaker cash-market and margin activity.
• Southbound net inflowUS$1.6bnNet inflow during July 16 to 22.
• Month-to-date southbound net inflowUS$11.1bnShowing continued capital support for Hong Kong stocks.
• Year-to-date southbound net inflowUS$46.6bnEquivalent to 46% of the level in the same period last year.
• Remaining budgeted fiscal quotaapproximately Rmb2trnThe report expects policy to be more likely to accelerate the implementation of existing quotas rather than launch a supplementary budget.
• June primary home salesdown 7% year-on-yearWeakened from 1% year-on-year growth in May, suggesting real estate may still weigh on domestic demand in the third quarter.

Impact & implications

The report's investment implication is that, in the short term, A-share sentiment may continue to be constrained by limited policy visibility, IPO supply pressure, and global AI-related volatility, making it difficult for risk appetite to expand quickly; by contrast, Hong Kong stocks have greater near-term allocation appeal due to earnings catalysts, AI commercialization, and capital flow factors. If the pace of policy from the Politburo meeting accelerates, earnings revisions stabilize, evidence of AI capital expenditure strengthens, and the global interest-rate path becomes clearer, the recovery rally in Chinese equities could become more sustainable.

Risks

• Policy response strength or timing may fall short of market expectations.
• The CXMT IPO and subsequent large IPOs may divert liquidity.
• Real estate data may remain weak, weighing on domestic demand in the third quarter.
• The breadth of earnings forecast revisions remains negative; if it cannot stabilize, it will weaken the sustainability of market recovery.
• Uncertainty over the global interest-rate path and the pace of Fed easing.
• Volatility in global AI-related sectors, deleveraging pressure, or a decline in risk appetite may weigh on Chinese equity performance.

What to watch

• The policy response from the July Politburo meeting to weaker second-quarter GDP and soft domestic demand.
• The implementation speed of the remaining approximately Rmb2trn in budgeted fiscal quota.
• The impact of the CXMT IPO and large IPO supply in July on market liquidity.
• Whether A-share turnover, ChiNext turnover, margin balances, and MSASI rebound again.
• Whether the breadth of earnings forecast revisions can stabilize and turn positive.
• Progress in AI commercialization, AI capital expenditure signals from major U.S. technology companies, and whether global tech-stock volatility eases.
• Whether southbound funds continue their net inflow trend.