PurposeThis study aims to explore three key questions: How does air pollution affect real estate prices across markets, considering income, demographics and urban density? How do spatial econometrics improve air pollution impact estimation compared to traditional hedonic pricing models (HPM)? What are the welfare implications of air pollution, particularly in terms of compensating surplus (CS), and how can a multistage HPM framework enhance economic loss assessment? By addressing these, the research aims to refine property valuation, inform urban policy and support sustainable housing and environmental regulations.Design/methodology/approachThis study uses a multistage HPM combined with spatial econometric techniques (Spatial Lag Model [SLM] and Spatial Error Model [SEM]) to estimate the impact of air pollution on housing prices in Thailand. To address endogeneity, a two-stage least squares (2SLS) approach is implemented using distance to factory and rainfall as instrumental variables. The marginal implicit price (MIP) and CS are derived to quantify economic losses. The Hausman and Hansen J-Tests validate model robustness. This integrated framework provides a spatially aware valuation of air pollution's economic burden, offering policy-relevant insights for sustainable urban and housing market management.FindingsThis study confirms that air pollution significantly reduces housing prices, with ordinary least squares overestimating the MIP due to ignored spatial dependencies. Spatial econometric models (SLM and SEM) and 2SLS provide more precise estimates, addressing endogeneity and valuation biases. The average MIP of air pollution per house is 39,838 THB ($1,138), while the CS averages 568,078 THB ($16,578) per household, highlighting substantial economic losses. The Hausman and Hansen J-Tests validate the robustness of the instrumental variables. These findings emphasize the importance of spatially aware valuation techniques in real estate pricing and environmental policymaking.Research limitations/implicationsThis study is limited by data availability and potential measurement errors in air pollution and housing price records, which may introduce estimation biases in Thailand. While spatial econometric models (SLM and SEM) and 2SLS help address spatial dependence and endogeneity, some unobserved market dynamics could still affect results. In addition, findings are specific to Thailand, limiting broader applicability. Future research should incorporate higher-resolution pollution data, alternative instruments, and longitudinal analysis to improve accuracy.Practical implicationsPolicymakers should encourage real estate developers to adopt green building standards and integrate air filtration systems into urban housing projects. Localized interventions, such as pollution taxation, congestion pricing and differentiated property tax schemes based on environmental quality, can help internalize the external costs of pollution and promote more sustainable urban development.Social implicationsIntegrate environmental justice principles into urban development to protect low-income communities that often reside in high-pollution zones. Incentivize development of sustainable housing projects with better air quality and urban greenery.Originality/valueThis study contributes to the literature by integrating spatial econometric models (SLM and SEM) and 2SLS to address spatial dependence and endogeneity in air pollution valuation. Unlike traditional HPMs, this approach provides less bias estimates of the MIP and CS, quantifying the economic burden of air pollution on housing markets. The findings offer policy-relevant insights for urban planning, environmental regulation and real estate valuation.
