关键词: 新基建, 经济效应, 能耗排放, 动态投入产出模型, 熵权TOPSIS法

Abstract: Based on Chinas noncompetitive inputoutput table and the Fourth National Economic Census data，this paper combined the structural Miyazawa multiplier and the dynamic inputoutput model with the endogenous “capitalincomeconsumption” relation of residents to evaluate and compare the economic effects，energy consumption, and carbon emissions of the new infrastructure investment plan and traditional infrastructure scenarios during the 14th Fiveyear Plan period The entropy weightTOPSIS method is employed to identify the main factors influencing the comprehensive efficiency of investment in seven categories of new infrastructure The results show that the economic effects of new infrastructure investment are relatively smaller than those of traditional infrastructure However，the energy consumption and carbon emission intensity of new infrastructure can be reduced to 048 tons of coal equivalent per ten thousand yuan and 097 tons of CO2 per ten thousand yuan respectively by 2025，achieving a 2563% reduction in CO2 emissions In terms of unit investment efficiency，the GDP multiplier of new infrastructure is slightly lower than that of traditional infrastructure （114 versus 133） However，the share of output driven by endogenous consumption is higher in the new infrastructure scenario，indicating that new infrastructure could further promote the transformation of growth drivers in China Benefiting from favorable energy consumption and carbon emission performance，the comprehensive efficiency levels of industrial Internet and artificial intelligence have reached 065 and 060 respectively，higher than the other five categories of new infrastructures，such as 5G and ultrahigh voltage Therefore，by identifying the buckets factors of seven categories of new infrastructure in terms of economy，energy，and carbon emissions，this paper can provide policy recommendations for the arrangement direction and optimization decision of new infrastructure investment

Key words: new infrastructure, economic effects, energy consumption and carbon emissions, dynamic inputoutput model, entropy weight TOPSIS method