Based on the data from China Family Panel Studies (CFPS) from 2010 to 2022, this paper measures the degree of housing inequality among urban households nationwide. It employs nonparametric kernel density estimation and spatial Markov chain methods to analyze the dynamic evolution characteristics of housing inequality across different regions, and utilizes the coefficient of variation and panel models to examine σ- convergence and β-convergence. The results show that: The overall level of housing inequality among urban households in China followed a trend of “rising first and then declining”, with gaps between major economic regions gradually narrowing; the characteristics of dynamic evolution in housing inequality vary significantly across regions, with provinces in coastal economic zones showing more pronounced polarization; σ-convergence is observed at the national level and within four of the eight major economic zones—the Eastern Coastal, Middle Yellow River, Middle Yangtze River and Southwest economic zones. Absolute β-convergence and conditional β-convergence are present in all zones except the Southern Coastal economic zone. Furthermore, the impacts of economic development level, housing prices, public expenditure, income level and urbanization level on housing inequality exhibit significant heterogeneity. Therefore, policy should focus on promoting the internal coordinated development within regions showing high polarization and disparity, disseminating successful experiences from converging regions while enhancing cross-regional collaboration, and implementing targeted measures according to the heterogeneous effects of influencing factors. These steps are crucial for fostering coordinated progress in housing development both nationally and within each economic zone.