Abstract: This study extends traditional growth accounting by incorporating data elements, developing a theoretical framework that examines how distortions in capital, labor, and data factor allocations affect total output. Our model specifically analyzes the relationships between resource misallocation, digital new quality productivity development, and economic performance. Using 2011-2016 China Tax Survey data, we measure resource misallocation and digital new quality productivity levels across three key sectors: (1) scientific research and technology services, (2) information technology services, and (3) selected high-end manufacturing industries. Our methodology decomposes output contributions while estimating single-factor distortion effects, output losses, and potential gains. Key findings reveal: First, capital demonstrates persistent over-allocation and volatility, while labor and data allocations remain relatively stable across all sectors. Second, digital new quality productivity exhibits instability, initially showing minor fluctuations before significant variations, with particularly strong performance in digitally transformed manufacturing sectors compared to core digital industries. Third, China's growth primarily stems from digital new quality productivity gains rather than factor inputs, though substantial negative impacts from factor distortions outweigh these productivity benefits. Fourth, factor misallocation causes annual output losses of 54.54%-66.09%, with potential gains of 119.98%-194.90% achievable through distortion mitigation. Notably, capital, labor, and data factors (controlling for externalities) significantly influence digital new quality productivity development. These results carry important policy implications for China's digital transformation, suggesting the need for: (1) factor market price mechanism reforms tailored to industry-specific conditions, (2) strategic development of digital new quality productivity, and (3) focused cultivation of emerging and future industries. Our findings particularly highlight the critical need to address factor market distortions that currently constrain productivity growth.

摘要： 通过将数据要素引入到经济增长核算模型中，拓展构建了一个包含资本、劳动力和数据三要素扭曲影响总产出变动的理论分析框架。基于2011—2016年中国税收调查数据，测算分析了科学研究和技术服务业、信息传输、软件和信息技术服务业和部分高端制造业的资源错配程度和数字新质生产力水平，并对总产出变动进行了贡献度分解，进而估计了单要素扭曲改变效应、总产出损失和潜在收益。研究发现，第一，资本要素基本处于过度配置状态且波动较大，劳动力和数据配置相对适度且波动较小，且三类行业资源误置困境表现类似。第二，在整体和行业两层面，数字新质生产力培育水平不稳定，呈现先小幅度波动、后大幅度下降和上升的变动特点。第三，我国经济增长主要依赖数字新质生产力发展驱动，其次依靠要素投入贡献，而份额效应和要素扭曲效应产生了较为严重的负面影响。第四，要素扭曲配置导致年度产出损失约为54.54%—66.09%，如果可以有效消除要素扭曲影响，则可产生约119.98%—194.90%的潜在经济收益。