An objective understanding of the current situation and influencing factors of rural green development in China is an important prerequisite for effective formulation making of green development policies. Based on the panel data of 31 provinces of China from the year 1997 to 2017, this paper constructs and measures the rural green development efficiency (RGDE) based on Driving-Force, Pressure, State, Influence, Response (DPSIR) model and super-efficiency slacks-based measure (SBM) model. The results show that, the overall RGDE in China is fluctuating and rising from 1997 to 2017, and there are some differences between and within regions. The RGDE in developed areas is higher than that in developing areas, and coastal areas are higher than that in inland areas. The entire country, eastern, central, and western regions show σ convergence, which indicates that the RGDE is getting better, but there is no absolute β convergence, that means there is no "catch-up effect" between regions, but the gradient divergence showed central > eastern > western. At the same time, there is no conditional β convergence; the initial RGDE has a positive impact on the growth rate of RGDE, and the financial self-sufficiency rate promotes the growth of RGDE of the whole country and the western region, but inhibits the improvement of RGDE of the eastern region. The per capita GDP, mechanization degree, and agricultural industrial structure in rural areas did not promote the growth of RGDE. Based on the results, this paper puts forward some policy suggestions, such as promoting the classified implementation of rural green development policies, strengthening the top-level design, optimizing the existing agricultural mechanism and system, and guiding and standardizing the farmers' green production behavior.
Keywords: China; Convergence; DPSIR framework; Rural green development efficiency (RGDE); Super-efficiency slacks-based measure (SBM).