Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Ethnic divergence in energy consumption is a phenomenon that threatens the potential gain in welfare that developing countries can achieve with strong economic growth and rise in income. However, the underlying mechanism preventing ethnic groups from accomplishing a successful transition to modern fuels is not yet well-understood, requiring an in-depth analysis of interaction effects between ethnicity and rise in income. The case of highly ethnically-diverse Vietnam offers an opportunity to examine the role of race on energy transition at the stage when grid electricity is available and income begins to rise. A methodological framework exposes the direct effect of ethnicity as well as the indirect, income-related effect of ethnicity on the ability of rural households to increase their electricity consumption. Using data from Vietnam's 2010 Household Living Standards Survey, feature selection is conducted with a machine learning technique, the least absolute shrinkage and selection operator (LASSO), which allows building robust models in the context of high-dimensionality. Furthermore, a mediation analysis complemented with a non-parametric bootstrap approach shows that income acts as a full mediator of ethnicity with respect to electricity consumption and as partial mediator for electricity expenditure. The results thus reveal a positive interaction effect between income and ethnicity, indicating different effects of rising incomes for Kinh and non-Kinh households, where Kinh are more likely than non-Kinh to translate extra income in higher electricity usage. Our results highlight the immanent need to identify and address non-income barriers that create ethnic disparities in the ability of poor, rural households to increase electricity use.