Jing Wu

We study the causal impact of property right uncertainty on prices and speculative activities in the housing market. We exploit a unique feature of residential housing markets in the Chinese city of Shenzhen, where otherwise comparable housing units with different property right protections are located adjacent to each other. Some have full property rights (FPR) protection with a 70-year leasehold while others only have limited property right (LPR) protections. Using detailed data of listings, we show that sales market values property right protections, but the rental market does not. Our estimates suggest that the market perceives that the chance that property right of the FPR might be in doubt at the end of their 70-year leasehold ranges from 7% to 21%, and that there is more than 50% chance that the LPR may not receive any legal protection in any future year. We also provide evidence that housing units with LPR protections are more prone to speculative activities measured by turnover rates and price volatility. Finally, we find that a public release of the new urban planning codes increases the listing prices and reduces the turnover rates and price volatility of the LPR properties relative to their matched FPR counterparts.

We examine the causal effect of uncertainty in property rights on housing prices and speculative behavior. We take advantage of a distinctive setting in Shenzhen, China, where neighboring residential units that are otherwise similar differ in the strength of their property rights protections. Some units have full property rights (FPR) with a 70-year leasehold, whereas others have only limited property rights (LPR) protections. Using detailed listing data, we find that the sales market prices these protections, while the rental market does not. Our estimates imply that the perceived probability that FPR rights could be questioned at the end of the 70-year term is between 7% and 21%, and that the probability the LPR receives no legal protection in any future year exceeds 50%. We further show that LPR units are more susceptible to speculation, as reflected in higher turnover and greater price volatility. Lastly, we show that the public release of new urban planning codes raises listing prices and lowers turnover and price volatility for LPR properties relative to matched FPR units.Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

We measure city-specific housing and land market wedges from developers’ optimization conditions across 93 Chinese cities and quantify their consequences in a dynamic spatial equilibrium model in which heterogeneous households choose location and homeownership under hukou restrictions and down-payment constraints. Developers are on average subsidized on housing sales but taxed on land purchases, and both distortions shrink with city size. Removing the wedges raises housing prices most steeply in lower-tier cities, depresses homeownership, and lowers average urban welfare substantially: the housing subsidies function as implicit, self-targeting transfers to renters and prospective buyers along the urbanization transition, which a budget-equivalent cash transfer can only partially replicate. Incumbent owners gain while aspiring buyers lose even more, and the largest measured distortion—the land wedge—proves nearly welfare-neutral: wedge dispersion alone would misstate even the sign of the welfare gains from reform.

Housing platforms now deploy generative AI to show buyers how a home might look after renovation. Unlike listing photos or VR tours, which reproduce a property's current state, AI-generated content depicts an unrealized future state whose market consequences are not well understood. Using proprietary data from a leading housing platform, we study AI-generated renovation visualization (AIRV), which renders plausible post-renovation interiors within the platform's VR interface. Our empirical analyses find that AIRV helps buyers imagine living in a home after renovation, with effects concentrated in listings whose post-renovation state is hardest to imagine, such as shell, never-occupied, and unfurnished units. Two opposing pathways drive these effects. Along the first, AIRV enhances perceived diagnosticity and increases onlineto-offline conversion, reflected in a higher showing-to-click ratio and shorter time to first offline showing. Along the second, AIRV raises buyers' pre-visit expectations, increasing the likelihood of expectation disconfirmation during in-person showings. The resulting bargaining frictions lengthen the post-showing phase and accelerate seller price discovery, as intensified buyer feedback prompts sellers to make larger list-price adjustments, leading to lower transaction prices for these hard-to-imagine listings. The expanded buyer pool and price corrections account for the higher sale probability. This work extends the literature by showing that AI-generated content's market effects propagate beyond the online stage and establishing a dual-pathway tension inherent in AI-generated future-state content, where the same content that accelerates online-tooffline conversion simultaneously introduces post-showing bargaining frictions that lower transaction prices. The findings inform platform design and governance of AIgenerated content in housing markets.

The intensive utilization of residential space is crucial to the transition to a carbon-neutral residential sector, although it has received limited attention in the literature. We develop a methodology to estimate the volume of unused housing in urban China, defined as dwelling units built and sold for at least two years but never occupied. By early 2021, 17.4% of the housing stock built in China during the first two decades of this century remained unused. The construction and operation of unused housing produce 55.81 million tons of carbon dioxide annually at the national level, accounting for 6.9% of the Chinese residential sector’s carbon emissions or 26.5% of the carbon emission reductions achieved by China’s primary ongoing residential decarbonization efforts. Cutting down the volume of unused dwelling units can contribute significantly to China’s decarbonization in 2021–2030.

This research investigates the profound influence of information friction on housing market outcomes in Beijing, China. Leveraging an extensive dataset of administrative micro-level housing resale transactions, we employ a boundary discontinuity design and difference-in-differences model to explore the consequences of prohibiting school-district-quality labels in the online listing information. The resulting surge in information friction during the housing search process yields fascinating insights. Our results show the prohibition leads to a 2.55% reduction in transaction prices and a 22.99% increase in the seller's time on the market for the houses corresponding to the key primary schools, in comparison to the other houses. The extended time on the market is primarily attributed to the heightened challenges that potential buyers face in finding their ideal dwellings. Moreover, the changes in buyers' search behavior and sellers' time on the market prompt the sellers to lower the listing prices to attract potential buyers and lead to lower transaction prices. Overall, the empirical evidence highlights the importance of information friction during the housing search process.

In this paper, we study the immediate impact of fertility relaxation on the gender wage gap. We explore a 2013 policy shock that relaxed the One-Child Policy in China. Using an employer-employee matched administrative data, we show that after the policy shock, the salary of female new hires is reduced by 2.7% relative to the salary of male new hires, equivalent to a 47% increase in the gender wage gap in the data. The effect is more pronounced for female new hires aged below 35, implying the salary reduction is related to concerns about the increasing fertility of females.

Subsidized credit supply is a widely used public policy tool to improve housing affordability in major economies. However, the subsidized credit supply would shift the demand curve up, which may drive up housing prices and undermine housing affordability. In this paper, we exploit a policy shock that increased the subsidized credit supply for housing units smaller than 90 square meters in a major city in China. Based on an administrative housing-resale-transaction dataset and a difference-in-differences approach, we show that for every dollar increase in the subsidized loan amount, the housing price increases by 0.12 USD on average. More importantly, the price effects are roughly equal among buyers using different financial resources, implying a strong redistribution effect of the subsidized credit supply among sellers and buyers with and without subsidized loans.

This article reveals an informal eldercare arrangement in China by focusing on how the value of housing owned by aging parents affects adult children's caregiving behavior. Using three waves of national survey data in 2011–2015, we find that when the potential housing‐bequest value increases, adult children from multi‐child households provide more emotional support for their parents, demonstrating the role of the intergenerational‐exchange mechanism in China's current eldercare arrangement. However, such an informal mechanism is weakened by the prevalent social norms of son preference and is also less effective among emerging only‐child households. Therefore, we call for more market‐oriented arrangements.