Do landscape amenities impact private housing rental prices? A hierarchical hedonic modeling approach based on semantic and sentimental analysis of online housing advertisements across five Chinese megacities

Abstract

Real estate premium associated with landscape amenities is a well-studied topic with a primary focus on housing prices. Presumably, the willingness-to-pay for landscape amenities should be very different between homeowners and tenants. Thus far, how landscape amenities affect residential rental prices is not well understood. This paper takes advantage of the big data of online housing advertisements to unravel how landscape amenities are capitalized into rental prices across five Chinese megacities (Beijing, Shanghai, Shenzhen, Hangzhou and Wuhan). Natural language processing, the latent Dirichlet allocation in particular, is first employed to semantically analyze the geo-textual advertisements. It reveals that ‘landscape amenities’ is a typical topic and ‘park’ is a typical component for housing advertisements in the five megacities. The lexicon-based sentimental analysis further shows that the strength of the sentiments associated with the ‘landscape amenities’ varies with cities. A series of hierarchical hedonic models based on the extracted semantic and sentimental aspects are then established for each megacity after segmenting the rental market into submarkets. The capitalization effect of landscape amenities is significant in Beijing, Hangzhou and Wuhan, while it is not significant in Shanghai and Shenzhen. Finally, variance decomposition analysis and marginal implicit price calculation unveil to what extent landscape amenities contribute to residential rental prices. Based on these findings, we discuss several major implications for urban planning. Our study unsettles the popular presumption that landscape amenities are key determinants of real estate values. It renews our understanding of the economic values of landscape amenities theoretically and methodologically.