A social network perspective on household cooking fuel transition: Evidence from China

This study aims to investigate the influence of social networks on household fuel transition. First, a k-means clustering model is employed to classify various types of social networks based on the structural and functional relationship characteristics. Second, a panel logit model is adopted to assess how social networks with distinctive properties affect cooking fuel transitions. Furthermore, the variations in social impacts among different fuel transition spans and modes are compared. Additionally, a heterogeneity analysis is carried out to identify how households respond to the differentiated impacts of social networks on the cooking fuel transition. The empirical findings demonstrate that the expansion of social networks speeds up the shift to cleaner cooking fuels. Compared to restricted social networks, family-oriented, outgoing and relatives-and-friends-open social networks significantly increase the probability of rural households adopting clean cooking fuels by 4.9, 7.0 and 9.3 percentage points, respectively. The fuel transition over time is frequently observed in households with outgoing, family-open and relatives-and-friends-open social networks, entailing a shift from firewood to gas and from firewood to electricity. Notably, as the duration of the transition extends, different social networks exhibit positive effects in a trend of initially rising and then gradually declining, typically reaching their peak during the 4-year interval fuel transition. Furthermore, the cooking fuel transition of high-income, small-scale households in rural China is worth more attention, as it is more sensitive to social networks with highly diverse ties and strongly stable structures. Finally, policy implications for promoting clean cooking transition from the perspective of social networks are proposed.