Market Distortion,Resource Misallocation And China's Total Factor Productivity

China's economy has been on a new stage with significantly lower growth rate.Many explanations for China's slow down,one of which is Total Factor Productivity(TFP).Aggregate TFP depends not only on the TFPs of individual production units,but also on how inputs are allocated across the production units.That is aggregate TFP can be low because inputs are misallocated across heterogeneous production units.Latest research shows that,China's R&D/GDP Ratio grows fast to 2.05%in 2015,exceeding the average level of OECD countries,while contribution of TFP to China's GDP growth is negative,-34.6%,indicating serious resource misallocation.This paper tries to investigate the sources,amount and operating channels of China's misallocation.To investigate topics on misallocation,there are two main approaches,which we will refer to as the direct approach and the indirect approach.The appealing feature of the direct approach just described is that it attempts to assess the importance of specific underlying sources in generating TFP effects via misallocation.However,there are many factors that one might reasonably imagine to be important sources of misallocation that are extremely difficult to measure directly,and we do not have any way of knowing how important the easily measurable factors relative to the no-so-easily measurable factors might be.The starting point for the indirect approach is to the obsearvation that any factors that create misallocation can be thought of as generating wedges in the first order conditions of establishment optimization problems.The indirect approach focus on the wedges rather than on the underlying sources of the wedges,obviously also not knowing the one factor's importance relative other factors.This paper tries to establish a model which has the features of both direct and indirect approach,to investigate the level effect,size effect and dynamic effect of misallocation on aggregate TFP.This framework separates one specific factor's effect or co-effect of two or more factors from the total misallocation.For Chinese issues,this paper studies ownership distortion,tax distortion,size distortion,firm-exit distortion and regional distortion,which generate misallocation and TFP loss.The main findings are as follow:Firstly,from a perspective of ownership distortion,this paper decompose aggregate TFP lose into three parts,misallocation within the SOEs group,misallocation within the non-SOEs group,and misallocation between the two group,and then we conduct a counterfactual analysis to find the best way of reallocation.We find that:(1)TFP distribution of SOEs are converging to that of non-SOEs,however,allocation within SOEs group are much less efficiency than that within non-SOEs group,which is an important point to explain why aggregate TFP of SOEs is lower than that of non-SOEs;(2)the weight of non-SOEs in China's manufacturing industries grows so fast and the allocation efficiency of non-SOEs decrease so slow that misallocation with non-SOEs group contributes most to total TFP lose;(3)reallocation within non-SOEs group can increase aggregate TFP by 21.7%in 2007,which is the best way of reallocation.Secondly,from a view of firm-exit distortion,Characteristics of Zombie firms in China and their impact on aggregate TFP are studied in this paper.China has more zombie firms weighted by assets and liabilities in 2011-2013 than in 2008-2009,while almost the same zombie firms when evenly weighted.Most zombie firms are non-state-owned since 2005,and are distributed in over-capacity industries.Such zombie firms hinder China's TFP growth.A 1%increase in zombie asset results in a 0.162%decrease in TFP growth rate.According to the dynamic OP TFP decomposition method proposed by Melitz&Polanec(2015),zombie firms impact TFP significantly negatively by technology progress channel and resource reallocation channel,significantly positively by firm entry channel,while not significantly by firm exit channel.Thirdly,zombie firms distort real tax abundance of non-zombie firms.Zombie firms pay no income tax because they are unprofitable;however,receive subsidies from the government,which increase financial stress in the context of China's tax-sharing system.Spending more with less income,local government tax non-zombie firms more.Empirical results show that,zombie firms in terms of assets proportion increase non-zombie firms' real income tax rate,and increase income tax rate more in provinces with more fiscal deficit.In addition,firms in industries with less zombie firms are taxed more.Fourthly,from a perspective of size distortion,this paper develops a model to investigate how misallocation changes the firm size distribution.The empirical results shows that,sizes of most Chinese firms are not efficient and state-owned firms tend to be bigger while nonstate-owned firms tend to be smaller than their underlying efficient sizes.The Chinese firm-size distributions of 35 industries are more concentrated than their efficient distributions respectively,which leads to less big firms and small firms,but more mid-size firms,thus reducing the market concentration.Misallocation results in deviation of size distribution from TFP distribution,high-TFP firms produce less while low-TFP firms produce more.Moreover,Chinese industrial firm-size distributions don't follow Zipf's law.Fifthly,from a view of regional distortion,this paper shows that the misallocation within provinces is far higher than the degree of that between provinces by decomposing total misallocation.The misallocation within and across regions both show a sustained downward trend from 1998 to 2007,and most of the provinces has improved.This paper also studies important factors that impact on the resources allocation.Empirical results show that financial market development and ownership structure reform reduce the misallocation,government intervention and market segment increase the misallocation,while the effects of labor mobility,FDI and foreign trade(exports and import)are not significant.