In the context of “carbon peak, carbon neutrality”, it is important to explore the spatial correlation network of carbon emission efficiency in the construction industry and its formation mechanism to promote regional synergistic carbon emission reduction. This paper analyzes the spatial correlation network of carbon emission efficiency in China’s construction industry and its formation mechanism through the use of the global super-efficiency EBM model, social network analysis, and QAP model. The results show that (1) the national construction industry’s overall carbon emission efficiency is steadily increasing, with a spatial distribution pattern of “high in the east and low in the west”. (2) The spatial correlation network shows a “core edge” pattern. Provinces such as Jiangsu, Zhejiang, Shanghai, Tianjin, and Shandong are at the center of the network of carbon emission efficiency in the construction industry, playing the role of “intermediary” and “bridge”. At the same time, the spatial correlation network is divided into four plates: “bidirectional spillover plate”, “main inflow plate”, “main outflow plate”, and “agent plate”. (3) Geographical proximity, regional economic differences, and urbanization differences have significant positive effects on the formation of a spatial correlation network. At the same time, the industrial agglomeration gap has a significant negative impact on the formation of such a network, while energy-saving technology level and labor productivity differences do not show any significant effect.

With the rapid economic development in recent years, China has increased its investment in infrastructure construction, and the construction industry has become a significant contributor to China’s carbon dioxide (CO2) emissions. Therefore, carbon emission reduction in the construction industry is crucial to achieving the goal of “carbon peaking and carbon neutrality” as soon as possible. However, few studies have investigated the factors influencing CO2 emissions from the construction industry in terms of spatial and temporal differences. To address this gap, we first improve the calculation method for the construction industry’s life-cycle assessment (LCA). The geographically and temporally weighted regression (GTWR) model is then utilized to provide insight into the spatio-temporal heterogeneity of the various factors influencing CO2 emissions across other regions and times. The results show that: 1) CO2 emissions from the construction industry in China increased rapidly from 576.5 million tons (Mt) in 2004–3,230 Mt in 2012 and then gradually decreased to 1998.51 Mt in 2020; indirect CO2 emissions accounted for more than 90% of the total CO2 emissions after 2008. 2) There is a solid global positive correlation between CO2 emissions from the construction industry in China during most of the time, and the spatial distribution of CO2 emissions shows a northeast-southwest pattern, with the center of gravity gradually shifting from central China to the southwest. 3) Economic output and industrial agglomeration are positive factors for the increase of CO2 emissions from the construction industry; and urbanization level, production efficiency, and energy efficiency are inhibiting factors for the increase of CO2 emissions from the construction industry. But the contribution and trend of each influencing factor differed significantly across time and regions, showing substantial spatial and temporal heterogeneity. Our findings provide a scientific basis for the Chinese government to implement a regional carbon reduction strategy for the construction industry.