中国制造业总量生产率增长的来源——基于PL方法的分析

The Source of Aggregate Productivity Growth in China Manufacturing

本文利用Petrin&Levinsohn(2011)(简称PL方法)提出的总量生产率增长的分析方法,通过对2000～2007年制造业总量生产率增长的分析表明,制造业总量生产率年均增长率为14.7%,其主要来源于存活企业的技术效率增长率和资源配置效应,而企业的进入和退出对总量生产率增长的贡献不大。对各省制造业的分析发现,各省区的总量生产率增长存在着显著的差异,其他地区增长率快于东部地区,这为近年来区域经济差异的缩小提供了一个可能的解释。而且,东部地区和其他地区制造业生产率的增长分别主要依赖于技术创新和资源有效配置。最后,生产率水平发生急剧变化的企业占制造业的比重不高,但对其总量生产率增长的影响很大,两者影响基本抵消;而生产率水平维持在高位的企业对总量生产率增长的贡献最大。

The paper uses the method proposed by Petrin and Levinsohn （2011） to analyze every Chinese manufacturing firm＇s contribution to the aggregate productivity growth of the whole manufacturing from 2000 to 2007, and decompose those contributions into within-firm allocation. Aggregate productivity growth is defined as the technical efficiency progress and between-firm resource difference between the change in aggregate final demand and the change in aggregate costs on primary inputs such as labor and capital, which links the demand side with the production side and therefore makes it have welfare implications. Using the decomposition methods on Chinese manufacturing, we mainly answer three questions ： （ 1 ） which is the main source of China growth, input accumula- tion or TFP growth？ （2） where does the TFP growth come from, the technology advancement or effective use of in- puts？ （3） is there significant difference in the productivity dynamics across provinces and among firms with different level or growth rate of productivity？ Answers to those questions help to understand of the Growth and provide a perspective about the role of technical progress and institutional changes cle. The data we use in the analysis come from the annual focus on the manufacturing industries. The analysis result growth rate of the value added of Chinese manufacturing is China＇s Pattern of growth mirasurvey conducted by National Bureau of Statistics and we shows that during the period of 2000 -2007 the average 15.27%, about 93% of which comes from the aggregate productivity growth. The estimate is high compared to existing research because of our different definition on aggregate productivity growth. Decomposing the aggregate productivity growth shows that on average the main source of aggregate productivity growth is from the incumbents rather than the net effect of entrants or exiters. That is, the incumbents contributed 89% of the aggregate productivity growth during the period while the net effect of exit and entry just contributed the rest 11%. For the incumbents, the technical efficiency makes almost the same contribution as resource allocation effect. However, it has a standard deviation that is 40% larger, which shows that the resource allocation has become a more stable support for the TFP growth in Chinese manufacturing than the technical progress. Then we analyze the respective contribution of labor, capital and intermediateinput to the resource allocation effect and find that the effective use of intermediate input makes the largest contribution each year during the period. The capital played a positive role in the aggregate productivity growth. However, its contribution is small compared to intermediate input, which might be due to the high adjustment costs associated with the capital reallo- cation. Contrary to the former two inputs, the labor input made a negative contribution to the aggregate productivity growth, indicating that labor input was not reallocated to more efficient firms. The reason might be that the rigid household registration system blocks the reallocation of labor across provinces. Therefore, policy makers should eliminatethe barriers to the labor movement. The analysis conducted on every province shows that for most provinces, the resource allocation effect is larger and has lower volatility than the technical efficiency progress, which is similar as the nationwide TFP growth dynamics. Nevertheless, significant difference in aggregate productivity growth is found among provinces. Specifically, the aggregate productivity growth of eastern provinces is lower than that of other provinces, which might explain why the region economic disparity is narrowing in recent years. Finally, I dis- cuss the role of firm heterogeneity in aggregate productivity growth in terms of the level and growth of productivity. The result shows that firms whoseproductivity move up or down rapidly contribute a lot to the aggregate productivity growth in opposite direction and their net effect is only 3 %. However, the firms that stay within the quintile of high productivity make the largest contribution to the aggregate productivity growth and it mainly comes from the resource reallocation.