Can digital finance promote inclusive growth to meet sustainable development in China? A machine learning approach

The global spread of the COVID-19 epidemic has caused increasingly grievous issues such as poverty, inequality and economic recession, which has hindered the realization of inclusive growth (IG) and disrupted the sustainable development trajectory. Meanwhile, with the vigorous development of digital finance (DF) based on advanced digital technologies such as big data, the Internet of things and artificial intelligence, new vitality has been injected into China’s growth model. Thus, whether DF could affect IG and to what extent has drawn attention from scholars to policymakers. This study examines whether DF significantly contributes to IG using the XGBoost machine learning (ML) algorithm for the first time. Using a panel of 281 prefecture-level cities from 2005 to 2020, we employ the Entropy-VIKOR model to assess cities’ inclusive growth index and reveal the spatial–temporal evolution and regional differences characteristics. We find that DF plays an indispensable role in promoting urban IG and influences the three sub-dimensions of IG: economic growth, opportunity equity and achievement sharing. The heterogeneous analyses based on geographic location and population size show that digital finance plays a more significant role in promoting inclusive growth of cities in central and western China than cities in eastern China; however, cities with different population sizes have little difference. Our findings using ML algorithms are robust to using traditional econometric models. This study sheds light on how DF could help achieve the IG in developing countries similar to China.