The direct share fell to 5.8% in 2025. This is the narrowest official measure. In 2024, the real estate sector’s value-added was ¥8.46 trillion, or 6.3% of GDP—already down from 6.8% at the start of the year, and a full half-percentage-point below 2023. By 2025, this had fallen below 6%. For context, the sector now contributes less to GDP than wholesale and retail (¥13.8 trillion), financial services (¥9.85 trillion), and construction (¥8.99 trillion). It is no longer the largest single industry, a status it held during the peak years of the property boom.
The broad contribution has fallen further in absolute terms. At its peak, real estate’s direct and indirect contribution approached 30% of GDP. Citi’s estimates put this combined figure at around 13% in 2025, while Bloomberg Economics measures the broader property-related share at 18.3%, down from 19.2% in 2024 and roughly 25% during 2015–2018. The divergence between these estimates reflects different methodologies—Citi’s narrower definition excludes some downstream effects that Bloomberg includes—but both agree on the direction and magnitude of the decline.
The investment channel remains the most visible transmission mechanism. Real estate development investment totaled ¥8.28 trillion in 2025, down 17.2% year-on-year . This contraction is large enough to drag down the entire fixed asset investment aggregate: total FAI (excluding rural households) fell 3.8% in 2025, but excluding real estate development, investment would have declined only 0.5%. In other words, real estate’s decline accounted for the overwhelming majority of the headline investment contraction. The sector accounts for roughly 20% of fixed asset investment, of which about 30% is land acquisition costs and 70% construction.
The policy implication is that real estate is no longer the engine it once was, but it remains large enough to matter. The 2026 Government Work Report places real estate under the framework of “preventing and resolving risks in key areas,” with the stated goal of “stabilizing the real estate market” through controlling new supply, destocking, and optimizing existing supply. The sector’s direct weight in GDP may continue to decline—Bloomberg projects it falling to roughly 17% on a broad basis by 2026, with high-tech and green industries surpassing it for the first time. But the transition is not costless: the 17.2% contraction in development investment in 2025 shows that the sector’s decline is still acting as a significant drag on the broader economy, even as its share of GDP shrinks.
2. Price-to-Income Ratio Comparison
The chart above ranks all 29 provincial-level jurisdictions by their price-to-income ratio in 2025. The methodology is consistent: it assumes a standard three-person household purchasing a 100-square-meter new home, and calculates how many years of disposable income would be required. The green dashed line marks the threshold below which housing is considered low-pressure (ratio under 6); the red dashed line marks the high-pressure threshold (ratio above 10).
Shanghai is the most unaffordable jurisdiction in the country. Its ratio of 16.9 means a standard household would need nearly 17 years of income to buy a home. Beijing follows at 13.9. The third-highest is not a first-tier city but Hainan at 11.1, reflecting the province’s tourism-driven property market and the influx of outside buyers that has pushed prices far beyond local income levels. Guangdong (9.5) and Tianjin (8.6) sit just below the high-pressure line.
The national picture has improved significantly. The national price-to-income ratio fell from 7.9 in 2020 to 5.9 in 2025—a decline of roughly 25% in five years. This is the result of two forces: housing prices fell about 4% over the period, while urban disposable incomes rose about 29%. The improvement is real, but it is concentrated in the middle and lower tiers of the distribution. The top-tier cities remain far above the national average.
Hunan is the least pressured province in the country. Its ratio of 3.9 reflects the long-standing low property prices in Changsha, the provincial capital, which has become something of an outlier among major Chinese cities for its affordable housing market. Inner Mongolia (4.3), Guizhou (4.1), and Xinjiang (4.4) are also in the low-pressure band, reflecting weaker demand and lower price levels in these regions.
The chart tells a story of convergence, but not uniformity. The national average has fallen toward the low-pressure band, and the majority of provinces now sit between 4 and 6. But the top five jurisdictions—Shanghai, Beijing, Hainan, Guangdong, and Tianjin—remain above or near the high-pressure threshold. The gap between Shanghai (16.9) and Hunan (3.9) is more than fourfold. This divergence is the defining feature of China’s property market in 2025: the national crisis has eased for most of the country, but the most expensive markets remain stubbornly unaffordable for ordinary households.
3. Price Trends in Major Cities
The chart above shows year-on-year new home price changes for the four tier-one cities against the 70-city average in December 2025. The pattern is not a uniform correction. It is a single outlier surrounded by decline.
Shanghai is the only major city where new home prices rose. Its +4.8% year-on-year increase made it the sole tier-one city to achieve positive growth in both year-on-year and month-on-month terms, with December’s +0.2% monthly gain extending a trend that produced a full-year cumulative increase of 5.7%. The driver is structural: Shanghai’s core districts continue to attract high-income buyers and premium “good house” projects, while policy easing in 2025—including purchase restriction relaxations and mortgage rate cuts—unlocked pent-up upgrade demand. The city’s new home market is increasingly segmented between luxury projects that sell well and mass-market units that do not.
Guangzhou and Shenzhen recorded the steepest declines. Guangzhou’s -4.8% and Shenzhen’s -4.4% were both deeper than the 70-city average of -2.7%. Guangzhou’s weakness is particularly pronounced: the city’s new home transaction volume fell 11% in 2025, and roughly 60% of new units now sell for under ¥30,000 per square meter—a record share that reflects a market shifting decisively toward lower-priced stock. Shenzhen’s decline follows a sharper correction in 2024 and reflects the ongoing adjustment in its previously overheated peripheral districts.
Beijing’s decline was milder but persistent. Its -2.4% year-on-year figure was slightly better than the national average, but the city’s 1-12 month average was down 3.9%, indicating that the weakness was sustained throughout the year rather than concentrated in a single month. The monthly decline of -0.4% in December was an improvement on November’s -0.5%, but not enough to change the trajectory.
The more alarming signal is in the second-hand market. While new home prices fell 2.7% nationally, second-hand prices fell 6.1%—more than twice the rate. In December 2025, not a single one of the 70 cities recorded a month-on-month increase in second-hand prices. Beijing led the decline at -1.3%, followed by Nanjing and Wuhan at -1.2%. The new home market is being held up by premium projects and policy support; the second-hand market, which reflects the real purchasing demand of existing homeowners, is still searching for a floor. This divergence matters because second-hand prices are tied directly to household wealth and consumption confidence. Until that market stabilizes, the broader property correction is not over.
4. Housing Inventory Levels
The chart above shows the destocking cycle—the number of months required to sell all existing new-home inventory at the current sales pace—across China’s city tiers in November 2025. The green line marks the 14-month threshold widely considered the upper limit of a healthy market. Every tier exceeds it, but the magnitude of the overhang varies enormously.
The national absolute stock is beginning to decline. At the end of 2025, commercial housing inventory stood at 766.32 million square meters, a decrease of 32.59 million square meters from the start of the year. This is the first meaningful reduction after four years of accumulation that saw the stock rise from 524 million square meters in early 2021 to nearly 800 million square meters by early 2025—a cumulative increase of over 52%. The reduction reflects deliberate policy: controlling new supply and accelerating the sale of completed projects.
But the destocking cycle has lengthened, not shortened. The national cycle reached 27.4 months in November 2025, up from 18.8 months in January 2023. This apparent contradiction—stock falling but the time to clear it rising—is explained by the collapse in sales velocity. New home sales fell 8.7% in 2025 to 881 million square meters. Inventory is shrinking because developers have stopped building, not because buyers have returned. In the fourth quarter, first-tier city destocking cycles broke through 23 months and second-tier cities reached a historic high of 26 months.
The divergence between city tiers is the defining feature. First-tier cities sit at 17.1 months—above the healthy threshold but manageable. Second-tier cities are at 22.6 months. Third- and fourth-tier cities are at 40.3 months, nearly three times the healthy limit. The root cause in smaller cities is structural: the demand boost from shantytown redevelopment has faded, while population outflow and weak industrial bases mean new demand is not materializing. Some individual cities are far worse: Nanjing’s cycle has exceeded 38 months, while a handful of smaller cities still clear in under 10.
The policy response is focused on the lower tiers. The 2026 policy framework explicitly links new housing starts and land supply to each city’s destocking cycle, giving municipal governments greater autonomy to tailor supply controls to local conditions. The acquisition of existing commercial housing for conversion into government-subsidized affordable housing is being accelerated as a primary mechanism for clearing inventory, particularly in cities with large unsold stock. For the first-tier and strong second-tier cities, the problem is less about excess stock than about price and confidence; for the third- and fourth-tier cities, the problem is that the buyers may simply not exist at current prices and population trajectories. The inventory overhang in smaller cities is not a cyclical problem that policy can wait out—it is a structural mismatch that will take years to resolve.
5. Developer Debt Crisis
The chart above captures the two sides of the developer debt crisis in 2025. On the left, the aggregate reduction achieved by listed developers and the restructuring completed by defaulted firms. On the right, the remaining pressure concentrated in the sector’s most exposed survivor.
The aggregate deleveraging is substantial. The 34 listed developers that had reported 2025 results by April 2026 reduced total liabilities by approximately ¥890 billion, bringing the sector total down to ¥6.47 trillion. This follows a similar reduction in 2024, meaning the listed developer universe has cut debt by roughly ¥2 trillion over two years. The composition of debt has also shifted: the median interest expense fell from nearly ¥800 million to about ¥600 million, and the weighted average borrowing cost for restructured firms like Country Garden dropped to 3.76%. The shift from “short-term, high-cost” to “long-term, low-cost” debt is the clearest structural improvement in the sector.
The default wave has been largely processed. As of late 2025, 21 defaulted developers had completed or received approval for debt restructuring and reorganization, with a combined restructuring scale of approximately ¥1.2 trillion. This brought nearly ¥2 trillion of interest-bearing debt into a “safe period”—meaning the immediate risk of disorderly default has been contained. The landmark cases were Sunac, which cleared its $9.55 billion offshore debt, and Country Garden, whose ¥127 billion offshore restructuring was approved in November 2025. Eight of China’s ten most indebted developers have largely put the offshore restructuring process behind them.
But the crisis has migrated to the survivors, not disappeared. Vanke, once China’s largest developer and long considered one of the safest, is now the focal point of remaining stress. Its 2025 net loss reached ¥88.56 billion, 79% wider than 2024, driven by high-priced land entering settlement and additional inventory impairments. More immediately, Vanke faced a failed bond extension vote in December 2025 for a ¥2 billion note, with creditors rejecting all three proposals. The company completed ¥33.2 billion of onshore bond repayments in 2025, but it still has ¥14.68 billion maturing in 2026, with ¥11.27 billion concentrated in April–July. Its cash-to-short-term-debt ratio has fallen below 0.5.
The 2026 outlook is one of declining pressure but unresolved fragility. Total developer bond maturities are projected to fall to ¥352 billion in 2026, down from ¥454 billion in 2025—a reduction of over ¥100 billion. By 2027, the sector is expected to reach a point where new issuance can broadly cover maturities. Commercial REITs are emerging as a new channel for developers to shift from heavy-asset to asset-light models. But these projections assume no major new defaults. Vanke’s situation shows that even a state-backed developer with a major shareholder (Shenzhen Metro) can find itself on the edge of default when sales collapse and refinancing channels narrow. The aggregate numbers are improving; the tail risk is not gone.
6. Policy Interventions and Effects
The chart above shows the scale of the two policy pillars that defined the 2025–2026 response. The white list financing mechanism addressed the supply side, ensuring that projects already sold but not yet built could be completed. The demand-side subsidy, introduced in September 2026, addressed affordability for first-time buyers. Together they represent a shift from supporting developers to supporting projects and households directly.
The white list became the primary tool for containing delivery risk. By late 2025, approved loans for white list projects exceeded ¥7 trillion, and more than 7.5 million pre-sold homes that had been at risk of non-delivery were completed and handed over. The mechanism operates at the project level rather than the developer level: a developer in distress can still obtain financing for its viable projects, isolating the good assets from the bad. This is a deliberate design choice, and it is why the white list has been more effective than earlier developer-level support measures. By early 2026, the mechanism was being extended to more cities and more project types, with regulators emphasizing “expanding coverage and improving effectiveness”.
Local governments used special bonds to attack the inventory overhang directly. In 2025, 26 provinces announced plans to use land reserve bonds to buy back idle land, with a total proposed scale exceeding ¥750 billion across more than 5,500 parcels. Actual issuance reached roughly ¥300 billion. The pace accelerated in 2026: first-quarter real estate-related special bond issuance reached ¥225 billion, up 42% year-on-year, with land reserve bonds accounting for ¥91.8 billion. The mechanism is straightforward: local governments borrow at low rates (Ningbo’s land reserve bond carried a 1.8% coupon) to buy land that developers cannot develop, removing it from the supply pipeline and easing the downward pressure on prices. The scale is still short of what was announced, but the direction is clear.
The demand-side subsidy is the most direct intervention of the entire policy cycle. Announced on September 29, 2026, and effective from October 1, the program provides a 1 percentage point annual interest subsidy for up to five years on new first-home commercial mortgages, capped at ¥1 million in loan principal per household. Eligibility is restricted to homes under 120 square meters and under ¥1.5 million in price. For a buyer with a ¥1 million mortgage, the subsidy is worth roughly ¥50,000 over five years—a direct, visible reduction in monthly payments. Early data suggests it is working: in the first month, searches for eligible homes rose 12.3% and viewing appointments increased 18.7%, with second-hand transactions in 20 key cities up 20% year-on-year in October.
The policy framework has also shifted from crisis management to institutional reconstruction. The “828 New Policy,” issued in August 2026, introduced three foundational reforms: a project company system to isolate project finances from developer balance sheets, a main bank system to coordinate creditor interests, and a completed-home sales system to gradually replace pre-sales with existing homes for sale. Beijing, Shanghai, and Guangzhou issued local implementation rules within weeks. The government has also formally acknowledged that China’s housing market has entered a “stock era”—second-hand transactions accounted for 50.4% of total housing transactions in the first half of 2026, surpassing new homes for the first time nationally. The policy implications are profound: the future is about managing existing stock, not financing new construction. The interventions of 2025–2026 were designed to stabilize the transition. Whether they succeed will depend on whether demand-side support can revive transaction volumes before the structural reforms fully take effect.
Strategic Implications and Outlook
The chart above assesses the three dimensions that will determine the market’s trajectory through 2027 and beyond. The scores are qualitative, but the ordering reflects a clear pattern in the 2026 evidence: the institutional framework for the new model is largely in place, demand-side support has produced measurable but fragile results, and developer balance-sheet repair remains the weakest link.
The policy framework is the most advanced dimension. The “828” reforms introduced the three foundational institutions—the project company system, the main bank system, and the completed-home sales system—that will govern the sector for the next decade. The 15th Five-Year Plan formally enshrines these reforms and sets the goal of “building a new development model for real estate”. The market has confirmed the structural shift: in the first half of 2026, second-hand transactions accounted for 50.4% of total housing transactions, surpassing new homes for the first time nationally. The policy infrastructure is no longer the constraint. The constraint is execution at the local level and the willingness of developers to adapt.
Demand stabilization is working, but the foundation is fragile. The September 2026 demand subsidy—a 1 percentage point interest subsidy on first-home mortgages up to ¥1 million, capped at homes under ¥1.5 million and 120 square meters—produced immediate results. In the first month, searches for eligible homes rose 12.3%, viewing appointments increased 18.7%, and second-hand transactions in 20 key cities rose 20% year-on-year. Shanghai became the first city to record positive year-on-year growth in second-hand prices since the correction began. But the subsidy is scheduled to expire after one year, and its impact is concentrated in the affordable segment. The market’s core problem—weak income expectations and subdued consumer confidence—remains unaddressed by a temporary interest-rate discount. The IMF has warned that “sluggishness in domestic consumption and weakening investment amid continued property sector strains” cannot be offset by export strength alone.
Developer balance-sheet repair is the weakest dimension. The aggregate deleveraging of listed developers has been substantial—¥890 billion in 2025 alone, with the 34 largest listed developers reducing total liabilities to ¥6.47 trillion. But the crisis has migrated to the survivors. Vanke, once the sector’s safest name, reported a ¥88.56 billion net loss in 2025 and faces ¥14.68 billion in onshore bond maturities in 2026, with ¥11.27 billion concentrated in April–July. Its cash-to-short-term-debt ratio has fallen below 0.5. The 828 reforms are designed to prevent future Vanke-style crises by isolating project finances and coordinating creditors through the main bank system, but they do not resolve the existing debt overhang. The transition from “restructuring” to “sustainable operations” is incomplete for a significant portion of the sector.
The outlook for 2027 is one of stabilization at a lower level, not recovery to the old normal. The IMF projects China’s GDP growth slowing to 4.0% in 2027, with the property sector remaining a drag. Reuters surveys project new home prices falling 3.5% in 2026 before stabilizing in 2027. The stock era is not a temporary phase; it is the new structural reality. The market of the next decade will be defined by transactions of existing homes, renovation and upgrading of aging stock, and the provision of housing services—not by the mass construction of new units that defined the previous two decades. The policy framework for this new model is in place. What remains unresolved is whether the demand-side support can bridge the transition, and whether the developers that survive can operate profitably in a market that no longer grows.
Sources
National and provincial statistics: National Bureau of Statistics of China, “Statistical Communiqué of the People’s Republic of China on the 2025 National Economic and Social Development” (February 2026); National Bureau of Statistics, “2025年全国房地产市场基本情况” (January 2026).
Price data: National Bureau of Statistics, “2025年12月份70个大中城市商品住宅销售价格变动情况” and accompanying interpretation by Wang Zhonghua, Chief Statistician of the NBS Urban Department (January 2026); People’s Daily coverage of the December 2025 70-city price index (January 2026).
Price-to-income ratio: Calculated from NBS provincial disposable income and residential price data; methodology follows the standard 100m² three-person household convention used in Chinese statistical commentary.
Developer financials: Vanke Enterprise Co., Ltd., 2025 Annual Audit Report (April 2026); Morningstar equity research on China Vanke (March 2026); Dow Jones / Morningstar coverage of Vanke’s 2025 results (March 2026).
Policy documents: Ministry of Housing and Urban-Rural Development and Ministry of Finance, demand subsidy announcement (September 2026); “828 New Policy” framework documents (August 2026); 15th Five-Year Plan real estate provisions.
International assessments: International Monetary Fund, China Article IV consultation and World Economic Outlook projections (2026).
Supplementary market data: China Index Academy, 100-city inventory and land reserve bond reports (2025–2026); CRIC and Wind data on listed developer liabilities.
Note on methodology: Where estimates from different research houses diverge (for example, Citi’s ~13% broad contribution vs. Bloomberg’s 18.3%), the report cites both and explains the methodological difference. The provincial price-to-income ratios are calculated at the provincial level and smooth out city-level extremes; city-level ratios for Shenzhen, Beijing, and Shanghai are substantially higher and are noted in the relevant section.