International Choice Modelling Conference, International Choice Modelling Conference 2017

Font Size: 
Choice experiment valuation of urban green spaces in Cape Town
Dambala Gelo Kutela, Jane Turpie

Last modified: 28 March 2017


Although many cities have guidelines on the quantity of green open space that should complement residential development, unfortunately there is little or no guidance on the type of these spaces in many of emerging countries. The present study used a choice experiment method for valuation of urban green space on the basis of its constituent attributes in Cape Town. We employed a combination of discrete choice models to evaluate preferences and preference heterogeneity for the following attributes; distance to small (community) parks and big (multi-use) parks, removal of litters, tree planting along streets and green belt development along river’s banks. Results from standard conditional logit, random parameter logit model and latent class logit models suggested that green belt, litter removal and street tree planting increases the demand for urban green spaces. Litter removal appears to offer the highest welfare gain compared to all other attributes. Interestingly, our analysis shows that proximity to both community park and multi-purpose parks are not preferred up to certain level of distance to these parks. This non-linear relationship points to security concern over the association between parks proximity and related crime incidence in one’s neighbourhood. Moreover, our results from random parameter logit model and latent class logit model respectively revealed evidence of considerable preference heterogeneity across individuals and six discrete classes of our study sample. We found that preference (latent) class membership is mainly driven by location of one’s residence within the city. Overall, our analysis suggest that urban green development that involve litter removal, greenbelt and street tree plantation in its design offers substantial welfare gain in our study context. However, preference heterogeneity across classes of our study sample underscores that such programs should be tailored-made on the basis of spatial clusters of the city.

Conference registration is required in order to view papers.