Does broadband infrastructure really affect consumption of rural households? – A quasi-natural experiment evidence from China

Purpose: As an important public infrastructure, broadband has absorbed a large amount of investment in China. However, how and to what extent these investments affect economic and social development is largely unknown. The purpose of this paper is to analyze the impact of broadband infrastructure construction on consumption of rural households, using an exogenous policy shock introduced by the China's “Broadband Countryside” pilot project. Design/methodology/approach: Using the tracking sample data of China Household Financial Survey in 2013 and 2015, this study estimates the effect of broadband construction on rural household consumption and draws causality between them relying on a quasi-natural experiment based on an exogenous policy shock. Findings: The difference-in-difference estimates show that broadband construction has significantly increased rural household consumption by 16.69%. This positive promotion effect is mainly achieved through mobile phone access to the Internet, while penetration of computer crowds out rural household consumption. Further research find that broadband construction has increased rural household consumption related to daily life and high-quality household consumption, but not statistically significant for the latter, and it has not helped to promote the consumption upgrading of rural households. Originality/value: This study contributes to the understanding of the positive status of broadband infrastructure in economic and social development by analyzing the impact of broadband infrastructure construction on rural household consumption. This study expands the content of consumption to rural households, especially high-quality consumption and consumption upgrading in rural areas, which provides the possibility to further tap the consumption potential of rural market. The study is the first to explore how broadband infrastructure construction affects consumption of rural households using a quasi-natural experiment.