Real estate prices in cities, especially in historic central areas, are influenced by the quality and character of their neighbourhoods. Many cities are identifiable by iconic historic features such as their buildings or monuments. This paper examines the impact of historic amenities on residential housing prices in the city of Lisbon, Portugal. Our study is directed towards identifying spatial variation of amenity values for churches, palaces, lithic (stone) architecture, and other amenities through the housing market. We develop a theoretical model to guide the empirical specification and establish a bid price function allowing for a spatial pattern of housing prices more complex than the monocentric city model. As such the willingness to pay may no longer be a monotonically decreasing function of distance to the CBD. A novel feature of our model is the introduction of herd behaviour in the housing market to capture underlying spatial interaction between property values. Empirically we estimate two global spatial hedonic models, a spatial lag and spatial error, to capture the effect of concentration of historic amenities on residential prices and correct spatial dependence. A locally weighted regression model investigates spatial non-stationarity and generates local estimates for individual historic landmarks. Results show significant spatial autocorrelation and there are benefits to modeling this behaviour through spatial hedonic models with reduced SSE up to 4% relative to OLS models. Direct proximity to any type of monument yield premiums of around 2% on housing prices however increased landmarks overall decrease prices by 0.7%, and different types of historic and individual landmark amenities induce varying housing premiums. Higher concentrations of non-landmark churches within 1,000 meters yield negative effects of 0.1% on housing prices with landmark churches having a larger impact around 4%. In contrast, lithic structures and palaces have positive effects with premiums in the order of 5% and 12% respectively and stronger effects for historic amenities located near open spaces. The impact of these historic amenities are assumed different since they are primarily aesthetic whereas churches provide services and a congregation point for the community. Our results highlight the capacity of the LWR model in explaining price differentials for proximity to individual landmarks. We see localized effects for being located near specific landmarks from approximately 130 ? to be further from the Castle and Church of Saint Anthony in downtown to 200 ? and 800 ? respectively to be located closer to the Palace of Necessidades and Monastery of Jeronimos - both characterized by large gardens. From a policy perspective, these findings highlight the importance of conceptualizing the amenity value not just in terms of structural characteristics but how those characteristics interact with or are conditioned by social, economic and other local contextual features.
展开▼
