The direction of technical change: a study based on the inter-provincial panel data of China

Technical change is usually biased and benefits some factors more than others; however, theliterature on the effect of technical progress is limited to the direction of technical change. For developing countries, whether technical change is biased towards particular factors is ofcentral importance. We apply the framework of Acemoglu, D. (2002, ‘Directed technicalchange’, The Review of Economic Studies, 69, 781–809) to develop explanations for thefollowing questions: What kind of production function is appropriate to special economies?If and to what degree technical change is biased towards particular factors? Based on interprovincialpanel data of China, this paper shows that: First, biased production functions aremore appropriate than neutral ones. Second, the elasticity of substitution between labour andcapital is less than a unity within the interval of 0.50 to 0.70, which indicates theasymmetric effect of technical progress on the productivity of capital and labour (a biasclearly in favour of capital). Third, the extents of bias in CES, VES, and CEEDx are 0.424,0.108, and 0.213, and more biased towards capital at 3.6%, 1.3%, and 2.2% (from 1978 to2010). This paper indicates that a crucial issue in China is how to change the direction of technical change and promote the productivity of labour.