China’s Q2 GDP Tracking Estimate Falls to 4.4%; Weak Domestic Demand Exacerbates 'K-Shaped' Divergence

Overview

This macroeconomic tracking report on China’s economy, prepared by Morgan Stanley’s Asia-Pacific economic team, uses the latest May activity data to lower the second-quarter actual GDP tracking estimate from last month’s 4.5% to 4.4%. The core finding is that China’s economy is increasingly characterized by ‘dual-speed’ or ‘K-shaped’ dynamics: the export-oriented production side remains resilient, while the consumption- and investment-driven domestic-demand side continues to weaken. The report argues that this divergent pattern heightens Beijing’s urgency to accelerate fiscal deployment in the third quarter, though policy priorities will still center on capital expenditures (capex) rather than consumer stimulus—particularly spending on initiatives under the ‘Six Networks’ umbrella, such as AI computing networks, data centers, and smart grids.

Core views

Demand side: Both consumption and investment are weakening, highlighting the shortcomings of domestic demand. In May, retail sales unexpectedly turned negative year-on-year at -0.6% (consensus: -0.5%), marking the first decline since 2022, despite an early boost from the ‘618’ e-commerce promotion. This reflects waning effects of the trade-in policy and ongoing pressure on household incomes from a softening labor market. Fixed asset investment (FAI) fell sharply by 2.7 percentage points year-on-year to -10.7% in May, with infrastructure investment being the main drag (year-on-year: -10.8%, year-to-date), indicating a lull in project pipelines after front-loaded activity in the first quarter. Manufacturing capex, while supported by robust exports, is constrained by widespread overcapacity, with only limited spillover into the computer and equipment sectors. Production side: Export-driven production remains relatively resilient, but structural divergence persists. Industrial value added grew slightly above expectations to 4.5% year-on-year (consensus: 4.3%), with midstream and upstream production up 0.7 percentage points to 6.4%, led by high-tech manufacturing. However, upstream mining and raw-materials production remained weak, suggesting that the growth momentum driven by external demand has not been evenly transmitted throughout the industrial system. Real estate: The recovery remains doubtful, with narrow improvements. Secondhand home sales have picked up in some Tier-1 cities, but new-home sales are slowing, mortgage lending is lackluster, and real estate development investment continues to plunge (-24.3%), indicating that the sector’s overall improvement is very limited and unlikely to serve as a reliable pillar for stabilizing domestic demand. Policy outlook: Fiscal acceleration in the third quarter, but ‘heavy on capex, light on consumption.’ With roughly 60% of the annual government-bond quota still unused, the July Politburo meeting is expected to call for a faster fiscal rollout rather than launching a new round of large-scale stimulus. Policy focus will remain on capital expenditures, especially within the ‘Six Networks’ initiative—AI computing networks, data centers, and smart grids—as geopolitical tensions underscore the urgency of energy security and technological self-reliance.

Analysis framework

The report employs a macroeconomic analytical framework of ‘activity-data tracking → GDP tracking estimate → policy implications.’ First, it dissects three key demand-side indicators—industrial value added, fixed asset investment, and retail sales—to identify structural divergences in economic performance. Next, it aggregates these subcomponents into a quarterly GDP tracking estimate and dynamically compares it with the previous-month forecast. Finally, based on the pace of fiscal spending (government-bond issuance) and policy signals (timing of Politburo meetings), it infers the direction and intensity of policy responses. The key to this approach lies in distinguishing between ‘production-side prosperity’ and ‘demand-side momentum,’ as well as identifying delays and bottlenecks in policy transmission.

Methodology notes

• Macroeconomic framework

  GDP Tracking Estimate

  Before the release of official quarterly GDP figures, institutions use this method to estimate the current quarter’s GDP growth rate in real time, based on published high-frequency activity data (industry, investment, consumption, etc.). It helps anticipate economic trends and adjust forecasts. In this report, Morgan Stanley lowered the Q2 tracking estimate from 4.5% to 4.4%, reflecting dynamic revisions following weaker May data.

• Industry/Industrial Analysis FrameworkUpstream–Midstream–Downstream Transmission in the Supply Chain

  Supply-Chain Upstream–Midstream–Downstream Transmission Analysis

  By observing differences in production conditions across various links in the supply chain, one can assess the effective reach of demand pull. This report finds that midstream and downstream production are relatively strong (6.4%), but upstream mining is weak, and the spillover of export-driven growth into domestic capital expenditures is limited to the computer and equipment sectors, indicating that overcapacity constrains broader transmission.

• Cycles and Economic Conditions FrameworkPeak-and-Trough Analysis

  Temporal Alignment of Policy Timing with Economic Turning Points

  This involves analyzing the correspondence between the timing of policy interventions and the position of the economic cycle. The report argues that after front-loading fiscal measures in the first quarter, a pipeline lull emerged in the second quarter, leading to a cliff-like drop in infrastructure investment. A renewed policy acceleration in the third quarter may create a ‘second bottom-up support,’ but its effectiveness will depend on the speed of fund disbursement and project execution efficiency.

Key data

• Q2 GDP Tracking Estimate4.4%YRevised down from last month’s 4.5%, slightly below the implied official target of 4.5%
• Industrial Value Added Year-on-Year4.5%Slightly above expectations (consensus: 4.3%), with midstream and upstream production up to 6.4%
• Fixed Asset Investment Month-on-Month-10.7%Down 2.7 percentage points from the previous month, with infrastructure investment as the main drag
• Retail Sales Year-on-Year-0.6%First negative reading since 2022, compared with the consensus expectation of -0.5%
• Infrastructure Investment Year-on-Year (Year-to-Date)-10.8%Down 7 percentage points from the previous month, reflecting a lull in project pipelines
• Real Estate Development Investment-24.3%Continuing deep decline, with both new-home sales and construction starts sluggish
• Proportion of Unused Annual Government Bond Quota~60%Provides room for fiscal acceleration in the third quarter

Impact & implications

The report concludes that the intensifying ‘K-shaped’ divergence in China’s economy means that relying solely on external demand and the production side cannot sustain recovery; domestic demand shortfalls are becoming the primary bottleneck. The combination of negative consumption and a slumping real estate market increases the urgency of policy responses, yet structural constraints limit policy options: amid local-debt pressures and fiscal-discipline requirements, large-scale consumer stimulus is unlikely, while capital expenditures on AI and energy transition under the ‘Six Networks’ initiative take priority. This policy orientation offers potential benefits to the technology-infrastructure, new-energy-grid, and data-center supply chains, but provides limited support for traditional consumption, real estate, and related industries. The report also cautions that if fiscal rollout falls short of expectations or the external-demand environment deteriorates, downside risks could further intensify.

Risks

• Fiscal rollout fails to meet expectations, with slow issuance of government bonds and delayed project implementation
• Deterioration of the external-demand environment, with slower export growth undermining production-side resilience
• Prolonged slump in the real estate market, dragging down related supply chains and local-government finances
• A persistently soft labor market continuing to suppress household consumption intentions and capabilities
• Escalating geopolitical tensions affecting energy security and progress toward technological self-reliance

What to watch

• The tone and deployment pace of fiscal policy at the July Politburo meeting
• The progress of government-bond issuance and the commencement of infrastructure projects
• The pace of capital-expenditure initiatives under the ‘Six Networks’ umbrella—AI computing networks, data centers, and smart grids
• Subsequent trends in retail-sales and employment data to gauge whether consumption remains subdued