The non-linearity of hospitals' proximity on property prices: experiences from Taipei, Taiwan

Given the semi-obnoxious characteristic of hospitals, either being right next to hospitals or being too farther away without easy medical access indicates inconvenience to residents. Quantile regression is applied to examine the potentially non-linear effects of hospital spline distance on quantiles of property prices in Taipei Metropolis, Taiwan. The conventional continuous distance (from property to hospital) showed consistent negative impact on property prices, implying hospitals as amenities as generally believed. Nevertheless, the splines of hospital distance demonstrated a non-linear effect on property prices: the positive effect of spline k 1 (0-500 m) on property prices indicates farther away from the hospital, higher the prices, possibly due to stronger negative externalities; the negative impact of spline k 2 (500-1000 m) on property prices implies the negative impact of hospitals wears off as the positive impact reveals itself gradually; thus, the property prices fall as the distance increases; splines k 5 (2000-2500 m) and k 6 (2500-3000 m) also demonstrated significant negative effects. In short, hospitals would only be highly evaluated in a 'close-but-not-too-close' geographic location. From urban planning perspective, hospitals, which are crucial in ageing societies, may reduce its externalities by creating spatial barriers such as scenic roads to keep distance from adjacent properties.