The paper uses carbon emissions and carbon sinks to build a carbon lock-in coefficient model, measuring the carbon lock-in degree of 30 provinces and cities in China from 2007 to 2016. Through building a spatial panel model, we test the key factors affecting carbon lock-in degree under spatial, temporal and bidirectional fixed effects, and in order to avoid the unstable estimation results by empirical research，we use the first-order differential GMM to solve the endogenous problem of variables . The results show that a phenomenon of obvious spatial spillover effect in China's provincial carbon lock-in. Urbanization, regional construction , transportation and fdi are the important factors ,which lead to the deepening of carbon lock-in. Innovation abilities and environmental regulations can slow down the deepening of carbon lock-in effectively.