Chinese Household Wealth Migrates from Real Estate to Financial Assets

Overview

This report by Goldman Sachs Asia economists focuses on the evolution of China's household balance sheet. The core finding is that against the backdrop of ongoing real estate adjustment and household deleveraging, the primary engine of Chinese household wealth accumulation is shifting from real estate to financial assets. The report constructs a quarterly tracker to bridge gaps caused by lags in official data and provides a ten-year outlook on the future path of Chinese household asset allocation based on historical experiences from Japan and the United States.

Core views

Asset Side: After peaking in early 2023, total household assets declined for approximately six quarters and have now largely stabilized at around RMB 730 trillion in recent quarters. Asset structure has undergone significant changes: real estate has shifted from being a growth engine to a drag, while financial assets (especially equities) have contributed more since the policy pivot in September 2024. Specifically, the share of real estate in total household assets fell from 67% in mid-2021 to 52% in Q1 2026; the share of cash and deposits rose from 16% to 25%; other financial assets (including equities and bonds) increased from 15% to 20%; among these, the share of directly held equities edged up from 5% to 6%. Liability Side: The household sector remains in a deleveraging phase, with both outstanding mortgage balances and short-term consumer loans declining. Although the household debt-to-GDP ratio (~59%) is not extreme relative to developed markets, the debt-to-disposable income ratio stands at a high ~140%, far exceeding that of major economies, indicating that the debt servicing burden remains heavy relative to household cash flows. Wealth Effect and Consumption: Real estate remains the dominant factor influencing household confidence and consumption, as it accounts for ~60% of household assets, whereas equities account for less than 10%. Over 90% of households own property, while only ~25% of Chinese adults participate in the equity market; consequently, weak housing prices have a greater impact on mass consumption, while gains from financial assets primarily benefit a narrower high-income group. Outlook and Reallocation: The report expects that as the wealth accumulation function of real estate weakens and deposit rates continue to decline, household savings will gradually migrate toward a broader range of financial assets. Drawing on post-housing crash experiences in Japan and the U.S., this transition is typically gradual and skewed toward wealthier demographics. Under the base case scenario, the report forecasts that by 2035, the share of real estate in household assets may drop from the current 52% to ~42%; the share of directly held equities may rise from 6% to 11% (assuming 10% annualized growth); the share of insurance may increase from 6% to ~10%; while the share of deposits remains relatively stable.

Analysis framework

The report employs a three-layer analytical framework: 'Construction of High-Frequency Tracker + Historical International Comparison + Scenario Forecasting'. Layer 1: Due to significant lags in household balance sheet data published by the PBOC and national think tanks, the authors independently constructed a quarterly tracker covering four main components: Financial Assets (integrating PBOC deposit data, NFRA insurance data, MOHRSS enterprise annuity data, CCDC bond and WMP data, AMAC public fund and PE data, and retail equity holdings estimated by the GS strategy team); Real Estate Value (estimating modern housing stock based on cumulative sales area and applying second-hand housing price series; non-modern and rural housing values are inferred using discounts and urban-rural ratios respectively); Automobile Assets (estimated based on CAAM sales volume and CADA average prices, applying a 10% straight-line depreciation method); and Liabilities (PBOC household loan data + MOHURD housing provident fund loan estimates). Layer 2: Through international comparisons with Japan in the 1990s and the U.S. post-2008, the report analyzes typical paths of household asset allocation following housing market crashes—transitions are usually slow, and recovery in risk asset participation is concentrated among high-wealth groups. Layer 3: Based on the aforementioned historical experiences and trend extrapolation, a ten-year base case scenario is constructed to quantify the potential scope for changes in various asset shares.

Methodology notes

• Industry/Sector Analysis FrameworkSupply-demand framework

  Supply-demand perspective in household balance sheet analysis

  In analyzing the impact of real estate on the wealth effect, the report implicitly applies a supply-demand framework: the supply side involves the accumulation and repricing of housing stock, while the demand side reflects household willingness and credit capacity to purchase homes; the relative changes between the two determine shifts in real estate's share of total assets.

• Cycle & Sentiment FrameworkSentiment Inflection Point Analysis

  Inflection point identification in asset price cycles

  The report uses high-frequency data tracking (new and second-hand home sales, 70-city housing price index) to assess whether the real estate market is approaching a price stabilization inflection point, which is critical for determining when the wealth effect might shift from a negative drag to neutral.

• Macroeconomic frameworkCredit/debt cycle

  Household sector leverage and deleveraging analysis

  The report measures household leverage using both Debt/GDP and Debt/Disposable Income metrics, noting that headline indicators may mask actual debt servicing pressure—when household income as a share of GDP is low (~45% in China vs. ~75% in the U.S.), Debt/GDP underestimates the true burden.

• Quantitative/Factor/Portfolio TheoryBeta/alpha analysis

  Beta exposure analysis in asset allocation

  The report views real estate as the core beta (systematic exposure) of Chinese household wealth, while equity allocation of less than 10% implies extremely low household wealth exposure to capital market beta, explaining why equity market rallies alone struggle to boost overall consumption.

Key data

• Recent Scale of Total Household Assets~RMB 730 TrillionPeaked in early 2023, declined for ~6 quarters, and has now stabilized.
• Share of Real Estate in Total Household Assets (Mid-2021 vs. Q1 2026)67% → 52%Peaked at highest share; remains the largest asset category.
• Share of Cash/Deposits in Total Household Assets (Mid-2021 vs. Q1 2026)16% → 25%Significant increase reflecting risk aversion and savings accumulation.
• Share of Other Financial Assets (incl. Equities & Bonds)15% → 20%Contribution increased following the Sep 2024 policy pivot.
• Share of Directly Held Equities5% → 6%Absolute level remains low, but showing marginal improvement.
• Household Debt/GDP Ratio~59% (Q3 2025)Down from peak of 61% in early 2024; lower than U.S. (~70%) and Japan (~62%).
• Household Debt/Disposable Income Ratio~140% (2025)Far higher than major economies; heavy debt servicing burden.
• 3-Year Time Deposit Rate1.25%Significant decline from 1.95% in early 2024 and 2.60% in early 2023.
• Proportion of Households Owning Property>90%Broad coverage of housing wealth effect.
• Proportion of Adults Participating in Equity Market~25%Equity wealth effect concentrated in a smaller demographic.
• Projected Share of Real Estate (2035)~42%Base case scenario assuming weak house price growth and slowing new home demand.
• Projected Share of Directly Held Equities (2035)~11%Assumes 10% annualized growth, broadly consistent with historical experience.
• Projected Share of Insurance (2035)~10%~RMB 6 trillion annual addition, including valuation changes and net inflows.

Impact & implications

The report argues that Chinese household asset allocation is in the early stages of a structural transformation with profound implications for the macroeconomy, capital markets, and financial institutions. For capital markets, even gradual reallocation could become a significant source of structural demand for equity and insurance markets. However, the pace of transition will be constrained by multiple factors: the recovery of resident confidence, labor market performance, the extent of deposit rate cuts, and whether capital markets can deliver attractive risk-adjusted returns. For consumption, given that the housing wealth effect far outweighs the equity wealth effect and property owners vastly outnumber equity investors, a recovery in financial assets is unlikely to fully offset the negative impact of falling house prices on consumption until real estate truly stabilizes. The report notes that if tier-1 and tier-2 city house prices achieve partial stabilization within the next one to two years (with Shenzhen and Shanghai potentially leading), and absent major negative shocks (such as significant AI-driven job displacement), the positive impact on household confidence, consumption, and capital markets could be substantial. For financial institutions, signs are emerging of a shift in deposits from the banking system to Non-Bank Financial Institutions (NBFIs), reflecting a migration of funds toward wealth management products, mutual funds, insurance, and equities. Continued declines in deposit rates amid policies protecting bank net interest margins will serve as a key catalyst driving this reallocation.

Risks

• Prolonged downturn in the real estate market extends negative wealth effects, suppressing consumption and confidence recovery.
• Household deleveraging exceeds expectations, with high debt/income ratios constraining credit demand and consumption capacity.
• Weak labor market and fragile resident confidence limit the effectiveness of policy easing and interest rate subsidies.
• International experience suggests asset reallocation is typically slow and skewed toward the wealthy, potentially delaying overall structural adjustment.
• Major negative shocks (e.g., significant AI-driven job displacement) could interrupt the repair of household balance sheets.

What to watch

• House price trends in tier-1 and tier-2 cities, particularly leading indicators in Shenzhen and Shanghai.
• Sustainability of high-frequency data on new and second-hand home sales.
• Changes in household loan growth and prepayment trends.
• Deposit movements in Non-Bank Financial Institutions (NBFIs) as a leading indicator of fund flows into wealth management, funds, insurance, and equities.
• Spread between 3-year and 1-year deposit rates, reflecting residents' motivation to seek higher yields.
• Changes in labor market conditions and household confidence indices.