MENGYAO CAO

Beijing Resale Housing Market (2020-2026)

Beijing Resale Price Indices Remain Below BaselineAmid Wide Local Gaps

Data through May 2026 | Index base: July 2020 = 100 | Published July 2026

01 / Executive Summary

Price Indices Remain Below Baseline,With No Broad-Based Recovery

  1. Both price indices remain below the July 2020 baseline

    In May 2026, the Transaction Price Index stood at 80.8 and the Lower-End Listing Price Index at 79.7, with July 2020 set to 100.

  2. Sample-wide averages obscure sizable local differences

    From July 2020 to May 2026, the median decline in the district-level Transaction Price Index across Dongcheng and Xicheng was 18.2 percentage points less severe than the median decline across Changping, Shunyi, Tongzhou and Daxing. Dongcheng and Xicheng recorded smaller cumulative declines than the four outer districts in this comparison, although variation within each district remained substantial.

  3. Recent observations do not yet establish a broad-based recovery

    Between November 2025 and May 2026, the Transaction Price Index decreased by 3.2%, while the Lower-End Listing Price Index decreased by 4.1%. Further observations are needed to determine whether the indices are stabilizing. The available data do not yet indicate a broad-based recovery in either price measure.

May 2026 Index Levels

80.8 Transaction / 79.7 Listing

Transaction Price Index / Lower-End Listing Price Index

District-Level Difference

18.2 percentage points

Difference between the median July 2020–May 2026 Transaction Price Index declines in Dongcheng and Xicheng and those in four outer districts

Current Interpretation

No sustained stabilization is yet evident

The available observations do not establish a broad-based recovery in the reported price indicators. This report does not include a forecasting model.

02 / Market Cycle

Both Price Indices Remain Well Below Their Peaks

Because the two indices share the same base period, their broad movements relative to July 2020 can be compared descriptively. However, they are based on different observations and should not be interpreted as matched listing-to-transaction price differences for the same properties. Differences between their levels may also reflect differences in sample composition and index construction.

Transaction and lower-end listing price indices

July 2020 = 100; data through May 2026

Transaction Price Index Lower-End Listing Price Index July 2020 baseline
The Transaction Price Index peaked at 111.9 in September 2022, while the Lower-End Listing Price Index peaked at 111.9 in July 2022. By May 2026, the Transaction Price Index had declined 27.8% from its peak, while the Lower-End Listing Price Index had declined 28.8% from its peak. A reading below 100 means only that the index is below its July 2020 level; it does not, by itself, indicate that prices declined in the most recent month or period.

03 / District-Level Divergence

Central Districts Have Higher Pricesand Smaller Declines

The map shows absolute sample transaction price levels per square metre for each covered district. The dumbbell chart shows changes in district-level price indicators relative to the July 2020 baseline. A higher price level does not necessarily imply a smaller decline from the baseline.

Sample transaction prices across 10 Beijing districts

RMB per square metre; darker shading indicates higher sample transaction prices.

The map covers the 10 districts for which transaction-price observations are available. The district values describe price levels in the available sample and may not be representative of every submarket or residential development within each district. Differences may also reflect variation in the types of properties represented in each district.

Cumulative changes in transaction and lower-end listing price indices by district

Percentage change from July 2020 to May 2026; farther left indicates a larger cumulative decline

Transaction Price Index Lower-End Listing Price Index
Cumulative declines in the Transaction Price Index were smaller in Dongcheng and Xicheng and larger in Tongzhou, Daxing and Shunyi. The district-level listing and transaction indices show different cumulative changes relative to July 2020. This endpoint comparison does not, by itself, establish how closely the two indices moved together over time. District-level results may not be representative of every submarket or residential development within each district.

04 / Spatial Gradient

Price Declines Tend to Deepen With Distance,but the Link Is Weak

Groups farther from the city center generally showed deeper price declines, although district, building age, property quality, transport access and sample composition may also affect the results.

Distribution of residential-development price declines by distance band

Segments show P25–P75; dots show medians

Spearman’s ρ = −0.15, calculated using ordered distance-band categories rather than exact distances

The median residential-development-level price change was −26.9% in the 0 to <8 km band, compared with −32.3% in the 8 to <15 km band and −33.6% in the 15 to <25 km band. The 40 km or more band contains only 49 observations, substantially fewer than the other groups, and should be interpreted with caution. Because the calculation uses ordered distance bands rather than exact distances, the coefficient should be interpreted as an approximate ordinal association.

05 / Submarket Profiles

High Transaction Activity Does Not GuaranteePrice Stability

The vertical divider marks the sample median for the Transaction Activity Composite Score, while the horizontal divider marks the sample median for the Price Stability Score. The quadrants indicate relative positions within the current sample. They are sample-relative categories, not industry benchmarks or assessments of overall submarket quality.

Transaction activity and price stability scores across 196 submarkets

Horizontal axis: Transaction Activity Composite Score; vertical axis: Price Stability Score

Activity median: 51.2; stability median: 49.3

One submarket with a comparatively large transaction sample is labeled in each quadrant for visual reference. The labeled submarkets are illustrative and are not necessarily the highest-scoring submarkets in their respective quadrants. Other submarkets can be inspected by mouse hover or keyboard focus. Because the source does not provide the complete weighting formulas, the scores are used only for relative comparison within the current sample and should not be interpreted as validated absolute measures.

06 / Within-Submarket Variation

Price Changes Vary WidelyWithin the Same Submarket

P90–P10 is the difference between the 90th and 10th percentiles of the residential-development-level indicator within the same submarket. A larger spread indicates greater variation within the submarket and means that the submarket average is less representative of all residential developments within it.

Top 10 submarkets with the widest within-submarket variation

Ranking among the 196 submarkets in the main analysis sample

Chaoyang Park, Haidian Beibu Xinqu and Ganjiakou have some of the largest P90–P10 spreads in the current sample. The source provides only the P90–P10 spread for each submarket, not the underlying P90 and P10 values.

07 / Interpretation Limits

The Data Describe Current Conditions,Not What Comes Next

Findings within the analyzed sample

  • In May 2026, both reported price indices were below the July 2020 baseline.
  • The reported price indicators varied substantially across districts, submarkets and residential developments within the same submarket.
  • The Transaction Price Index and Lower-End Listing Price Index showed different cumulative changes relative to July 2020.

Current observations

  • Both sample-wide indices were lower in May 2026 than six months earlier.
  • Cumulative declines in the Transaction Price Index were smaller in Dongcheng and Xicheng than in the four outer districts used in this comparison.
  • Median declines tended to be larger in farther distance bands, but the ordinal rank correlation was weak (Spearman’s ρ = −0.15).

What remains to be tested

  • Whether the price indices stabilize across several consecutive monthly observations.
  • Whether the Lower-End Listing Price Index shows sustained stabilization in subsequent months.
  • Whether any future stabilization is broad-based across districts and submarkets rather than concentrated in a limited number of observations.

08 / Data & Methods

How the Sample and Metrics Were Constructed

Sample selection

Of the 7,652 residential developments in the initial sample, 571 lacked valid coordinates. A further 107 had coordinates but could not be matched to transaction data. After matching, 1,205 residential developments that did not meet the requirements for residential-development-level analysis were excluded, leaving a final sample of 5,769.

Metric definitions

Metric Meaning in this report How to interpret it
Transaction Price Index An index, normalized to 100 in July 2020, that measures changes in the observed transaction-price level relative to the base period. A reading above 100 indicates an observed transaction-price level above July 2020; a reading below 100 indicates a level below July 2020.
Lower-End Listing Price Index An index, normalized to 100 in July 2020, that tracks changes in the lower end of observed listing prices relative to the base period. A reading above 100 indicates that the lower-end listing-price level is above its July 2020 level; a reading below 100 indicates that it is below that level.
Transaction Activity Composite Score A composite score used for relative comparison of transaction activity across the 196 submarkets in the current sample. A higher score indicates stronger measured transaction activity relative to other submarkets in the current sample; it is not an absolute measure of market liquidity.
Price Stability Score A composite score used for relative comparison of measured price stability across the 196 submarkets in the current sample. A higher score indicates greater measured price stability relative to other submarkets in the current sample; it does not necessarily indicate rising prices or lower investment risk.

The complete construction formulas for the Transaction Activity Composite Score and Price Stability Score are unavailable. The scores are therefore treated as descriptive source variables and are used only for relative comparison within the current sample. They should not be interpreted as validated absolute measures, industry benchmarks, absolute ratings or investment advice.

Data Source

Sample transaction and listing data for Beijing’s resale housing market, sourced from online resale-housing platforms.

Disclaimer

This report is intended to help readers understand the trends, geographic differences and sample composition observed in the covered Beijing resale-housing data. It does not forecast short-term price movements and should not be interpreted as investment, valuation or transaction advice.