How do varying socio-economic factors affect the scale of land transfer? Evidence from 287 cities in China

With the rapid development of China’s social economy, the scale of land transfer has also increased, which has led to a new pattern of urban land space. This article uses global regression of ordinary least squares (OLS), spatial lag model (SLM), spatial error regression model (SEM) and local regression of geographically weighted regression model (GWR), and multi-scale geographically weighted regression model (MGWR) to explore the influence of socio-economic factors on the scale of land transfer. The relationship between these factors and the scale of land transfer varies greatly from region to region. The local model (MGWR) can express the non-stationary relationship between variables, and the regression estimation results are more robust. The results show that total investment in fixed assets (TIFA) and the non-agricultural population (NAP) had significant effects on the scale of land transfer in 2005, with regression coefficients of 0.964 and -0.247, respectively. In 2010, per capita GDP (PCG), population density (PD), proportion of tertiary industry in GDP (PTIG), and TIFA had significant impacts on the scale of land transfer, and the corresponding impact coefficients were 0.413, -0.085, -0.081, and 0.322. In 2015, the variable of PCG had significant impact on land transfer, with the coefficient of 0.048. The influencing factors of the scale of land transfer are changing at different points in time, and the formulation of land transfer policies should be treated differently according to the different socio-economic conditions in each period.