Accounting for China's growth in 1952 - 2008 : China's growth performance debate revisited with a newly constructed data set

Using adata fundamentalist approach,ʺ this study revisits the long debate about Chinas growth performance by seriously tackling the existing data problems that have been the major obstacles to a proper assessment of Chinas growth performance. First, this study examines and adjusts the serious break in the official employment statistics in 1990. Second, it provides an adjustment for the numbers employed by a human capital effect. Third, it tests the sensitivity of Maddisons (1998a) zero labor productivity growthʺ assumption in gauging the real growth of the so-called non-material (including non-market) services.ʺ Fourth, it further improves the authors earlier physical output-based production index for the industrial sector (Wu, 2002a) by using multiple weights and time-variant value added ratios obtained from the Chinese input-output tables. The likely problem of product qualityʺ in such a physical measure is examined and rejected. Fifth, it provides a new set of estimates of capital stock for the aggregate economy using alternative deflators and depreciation rates, crosschecked by the authors industry-level capital stock estimates (Wu, 2008b). This completely new data set is used in a Solow-type growth accounting exercise with different factor income share assumptions. The new resultsunder the full adjustment scenario for the post-reform period using input-output table income weightsshow that the estimated annual TFP growth rate is 0.3 percent, which is substantially lower than the estimate of 3.1 percent derived from the official data without any major adjustment. A range of TFP estimates is also provided for each sub-period under different assumptions.