Preference heterogeneity and cultural sensitivity : a comparative study on urban river restoration using discrete choice experiments

Abstract

Urban rivers provide ecological and social benefits (which are defined as ecosystem services) to urban residents. Yet, in both developed and developing countries, rivers in densely-populated urban settings have been significantly transformed and contaminated due to intensive human activities, and thus their ecosystem services have been notably degraded. Urban river restoration has, therefore, become a worldwide focus. An accurate understanding of residents’ preferences for urban river restoration, which may be partly dependent on cultural contexts, is crucial for forging public support for and participation in river restoration projects. The present comparative study aims at exploring differences in resident preference heterogeneity (potentially sensitive to cultural contexts) between China and Belgium. Discrete choice experiments were applied to estimate residents’ preferences for four river-related attributes including natural habitat, morphological feature, water quality, and recreational facilities and probe the impacts of environmental concerns and perception of ecosystem services on the willingness to support urban river restoration. An ordinal four-level scale was initially adopted to explore the effects of attribute non-attendance. Questionnaire surveys were conducted in Guangzhou (China) and Brussels (Belgium). The results show that Guangzhou residents have greater biospheric-egoistic- altruistic concerns than Brussels residents. More specifically, both groups of respondents are predominantly egoistic. Regarding the key ecosystem services of urban rivers, Brussels residents attach more importance than Guangzhou residents. The mixed logit models with correlations between random parameters of the attributes were applied for preference estimation. The results reveal that Guangzhou residents are apt to evaluate the attributes jointly (especially the provision of recreational facilities), whereas Brussels residents tend to evaluate the attributes relatively individually. Moreover, Guangzhou people’s support for restoring urban rivers may be mainly driven by their environmental (both biospheric and altruistic) concerns. In comparison, Brussels people’s support tends to be motivated by their perceived importance of urban rivers’ specific ecosystem services. The multinomial logit models with separate coefficients for different attribute attendance levels are implemented depending on the application of the polytomous scale for stated attribute attendance. Guangzhou residents exhibit notably higher degrees of non- attendance to natural habitat, morphological feature, and cost than their counterparts from Brussels. Regarding willingness-to-pay amount, water quality improvement is the dominant part of utility obtained from urban river restoration for two groups of residents. The attribute of morphological feature is the least valued in Guangzhou, whereas the provision of recreational facilities is the least in Brussels. For water quality improvement and recreation facility provision, Guangzhou residents are willing to pay more than Brussels ones. The pioneer comparative study helps better understand resident preference heterogeneity for urban river restoration in China and Belgium and the impacts of cultural contexts. Involving cultural contexts into choice experiments for environmental goods may contribute to explaining preference variation. Moreover, the study provides practical and instructive implications for attribute prioritization and cost-benefit analysis for urban river restoration projects and demonstrates directional ways towards forging better public involvement in the two countries.