Purpose This paper aims to develop and test a governance-sensitive structural model of housing purchase intention under urban flood risk in China. It examines whether the canonical disclosure-deterrence relationship documented in mature property markets travels to the dense, platform-mediated Chinese urban housing context and identifies the mechanisms through which flood-risk information is processed in a market dominated by asset-logic decision rules. Design/methodology/approach Primary survey data from 505 urban residents across 29 Chinese provinces are analysed using covariance-based structural equation modelling with maximum-likelihood-robust estimation. The model integrates eight latent constructs across three theoretical strata: distal cognitive drivers optimism bias (OB), flood information transparency, mid-level mechanisms housing bullishness (HB), perceived flood risk (PFR), community convenience and the proximal theory of planned behaviour. Multigroup analyses compare recent residents with long-term residents and renters with owner-occupiers. Findings Three results reframe the literature. Firstly, perceived information transparency exerts a positive effect on purchase intention indirectly through HB and a negative direct effect on PFR, supporting an information-abundance paradox. Secondly, an asset-mediated risk-absorption mechanism, operating through HB, OB and community amenity premia, is associated with weaker classical risk-deterrence. Thirdly, the canonical attitude-to-intention path collapses to non-significance once subjective norm, perceived behavioural control and the asset-logic block are jointly specified, with subjective norm dominating intention formation. Originality/value The paper introduces an information-abundance paradox and an asset-mediated risk-absorption mechanism to housing-market analysis, documents the attitude-transmission null in Chinese-urban applied theory of planned behaviour, and provides benchmark coefficients from a pan-Chinese sample.
Lin et al. (Sun,) studied this question.