Modeling heterogeneous vehicle ownership in China: A case study based on the Chinese national survey

Understanding the heterogeneity of vehicle ownership decisions in China is vital to accurately estimating the rate of vehicle ownership in its various provinces. In this study, we employ a latent class model to investigate the heterogeneity of vehicle ownership behavior, based on the China Household Finance Survey (CHFS) data. The results show that the households within the CHFS data can be categorized into two classes, and that households within each class rank the importance of socioeconomic variables in significantly different ways. For instance, with regards to deciding to own a vehicle, the households in Class 1 (the income-based class) rank household income as the most important factor, while the households in Class 2 (the comprehensive considerations class) rank household income, household status, and household size as being almost equally important. Further, the model coefficients also reveal the evolution of vehicle ownership in the near future, and how changes in macroeconomic variables may influence household vehicle ownership decisions. In application, the results can be used to assist policy makers in designing policies that control excessively high levels of vehicle ownership; they can also be used to help auto manufacturers pinpoint specific vehicle models to be sold in different regions of China, so as to drive the highest possible profits.