Export behaviour and firm productivity in China

Abstract

This study examines whether exporters become more productive through estimating a production function using Chinese firm-level data from 1997 to 2000. The results indicate that exporters are more productive and, more importantly, that export strategies promote the productivity of these firms. The test for the relative timing of export behaviour and gains in productivity provides strong evidence for the existence of learning-by-exporting effects rather than self-selection effects. These results are robust when controlling simultaneous bias and selection bias by using a fixed-effects model, SYS-GMM estimation and semi-parametric estimation.

Keywords:

• export
• self-selection
• learning-by-exporting
• estimation bias

Acknowledgement

The authors acknowledge the support provided by the National Social Science Foundation of PRC (07&ZD017), the National Soft Science Foundation of Ministry of Science and Technology of PRC (2008GXS1B022) and the ‘985’ Project of ‘Economic Openness and Trade Development’. The authors wish to thank the anonymous referee for very valuable comments. We also thank World Bank for providing the data. All remaining errors are our own.

Notes

Notes

1. China's Government Work Report (2008).

2. Clerides, Lach, and Tybout (1998) first introduced the self-selection effect. The range of extra costs, including transportation costs, distribution costs, marketing costs, costs related to hiring personnel with skills to manage foreign networks, and production costs in modifying domestic products for foreign consumption provide an entry barrier that less successful firms cannot overcome.

3. The samples in this paper come from five representative cities in China, so one should be cautious when extending the results of this paper to all firms.

4. We transferred the original chain price indices into fixed base-period price indices for convenience.

5. Non-exporters are firms that never exported goods during the sample period. Exporters are firms that exported throughout the sample period. Entrants are firms that began as non-exporters but began exporting during the sample period and then never stopped. Quitters are firms that began the period as exporters but ceased exporting during the sample period and never resumed. Switchers are firms that switched exporting status more than once during the sample period.

6.  are predetermined variables, that is, the interferences of the current variables affect this latter value .

7. In order to interpret the coefficient in Equation (4), we rewrite it as follows: with .

8. A full description of the Olley-Pakes estimation is beyond the scope of this paper and thus is only briefly outlined as the variables of interest in this study are the coefficients for exports and inputs. For a full description, please refer to Olley and Pakes (1996).

9. When we used the Olley-Pakes method, the variables of and are neglected in order to get the high-order polynomial of the logarithm of investment and capital. Blalock and Gertler (2004) adopted the same way when they used the Olley-Pakes method.