Dances with Chinese data: are the reform period Chinese provincial panel data reliable?

There is a debate over the reliability of the Chinese data (e.g., Young, 2003; Holz, 2003, 2006). In this paper we test the Chinese provincial panel data for the period 1978-2002 against the predictions from the technology diffusion model. We find that the estimated coefficient on initial real GDP per worker is negative and significant, showing strong evidence of conditional convergence; the estimated coefficients on secondary school human capital investment rate and labor force growth are positive and negative respectively, significant at the 5% level, in both LSDV (Least squares dummy variables) estimation and system GMM (Generalized method of moments) estimation that overcomes the endogeneity of these variables. The test accepts that the estimate coefficients on physical capital and human capital investment rates are equal, with absolute magnitudes about half of that on labor force growth in LSDV estimation. The estimated coefficient on the FDI to GDP ratio that captures technology diffusion is insignificant in LSDV estimation but becomes significant in system GMM estimation. All these are consistent with the technology diffusion model (and the augmented Solow model). Therefore, the reform period Chinese provincial panel data may be reliable for growth regressions.