Does development zone have spillover effect in China?

Abstract

We construct a new database of spatial relations between industrial firms and development zones using the Application Programming Interface addressing and document the development zones' spillover effect on firms' total factor productivity (TFP). Development zones significantly improve the TFP of surrounding firms, which attenuates with the distance from a development zone, and increases with the density of development zones in a given area. The spillover effect of national development zones decays more sharply than that of provincial ones. The rectification project improves firms' TFP and the spillover effect of development zones. The establishment of development zones affects surrounding firms' TFP, where TFP of newly entered firms decreases in the short term but soon rises rapidly.

Keywords:

• development zones
• spillover effect
• data mining
• Geographic Information System (GIS) analysis
• Propensity Score Matching and Difference-in-Differences test (PSMDID)

JEL classifications:

• L60
• O25
• R58

Acknowledgements

We are grateful to Professor Jun Zhang and Professor Ming Lu for their very helpful comments and suggestions.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. For more details, please refer to 2013 Research Report on the Investment, Transformation, and Upgrading of China Development Zones (in Chinese), by CEF (China's Economic Future). Analysis of the Trend and Development Report on China Industrial Park 2014 (in Chinese), by CEC (China Economic Consultants), also provides valuable information.

2. National Development and Reform Commission's Notice on Key Points in the Rectification of Different Types of Development Zones Report (Fa-Gai-Wai-Zi (NDRC Foreign Investment) [2003] No. 591, in Chinese), issued by National Development and Reform Commission (NDRC), Ministry of Land and Resources and Ministry of Construction.

3. We do not have the data of illegal zones throughout China. We collect manually the data of illegal zones in Guangdong province. In Section 5, we use this manually collected data of Guangdong province to discuss the sample selection problem.

4. General Office of the State Council's Urgent Notice on Further Management of Land Market Order (Guo Fa Ming Dian [2003] No 7. in Chinese).

5. General Office of the State Council's Notice on Rectification of Development Zones and Administration of Construction Land (Guo Ban Fa [2003] No. 70. in Chinese). General Office of the State Council's Urgent Notice on Further Management of Land Market Order (Guo Fa Ming Dian [2003] No 7. in Chinese). Notice on Rectification of Specific Standards and Policy Limitation (Fa Gai Wai Zi [2003] No. 2343. in Chinese).

6. Map data encoding and deviation refers to a migration algorithm that adds or subtracts a shift to longitude or latitude, resulting in a spatial mismatch with the real map.

7. Two density curves have passed the variance test before comparison.

8. We select firm's addresses (including detailed registered address, residential district, resident's committee, and the village or town) as the matching fields to estimate the actual radius of a development zone.

9. The regression in Table 4 shows results when the spillover radius is 10 km, while we test the robustness with the spillover radius as 5 km.

10. The coastal areas include Liaoning, Beijing, Tianjing, Hebei, Shandong, Jiangsu,Shanghai, Zhejiang, Fujian, Guangdong, Guangxi, and Hainan, the central areas include Heilongjiang, Jilin, Inner Mongolia, Shanxi, Henan, Anhui, Jiangxi, Hubei, and Hunan, and the western areas include Chongqing, Sichuan, Guizhou, Yunnan, Tibet, Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang.

11. As national economic and technological zone, the export processing zone and the bonded zone share high similarity in the administrative level, development path and function, the above three types were categorized as national economic zones.

12. As shown in column (5), . That is to say, if the distance between firm and nearest development zone is more than km, existence of four or above development zones would result in the dominance of the competition effect, which indicates an ‘over-dense’ phenomenon.

13. National Development and Reform Commission's Notice on Key Points in the Rectification of Different Types of Development Zones Report (Fa-Gai-Wai-Zi (NDRC Foreign Investment) [2003] No. 591, in Chinese), issued by National Development and Reform Commission (NDRC), Ministry of Land and Resources and Ministry of Construction.

14. According to Notice on Cancelling Newly Added 397 Development Zones like City Economic and Technological Zones (Yue Go [2004] No. 40, in Chinese), issued by General Office of Guangdong province: ‘the provincial government decides to cancel 397 newly established development zones, mainly prefecture- and county-level economic and technological development zones…For those closed zones, local governments shall cancel their names, administrative offices and withdraw administrative authorities.’

15. If a firm is in the estimated radius of illegitimate zones in 2002 or 2003, then it is categorized into the treatment group, which is under direct influence of the rectification policy. In Table 5, the estimated radius is 10 km and in the robust test, the radius was replaced as 5 km. No significant change was observed in the estimations.

16. If the distance from the firm is smaller than ‘the estimated radius’, it is defined as the in-zone firm.

17. Three types of national zones were under study: the national economic and technological development zone, the export processing zone and the bonded zone.

18. As fewest development zones established in 2005, the year was selected in the falsification test.

Additional information

Funding

Deming Luo acknowledges the financial support from the Natural Science Foundation of Zhejiang [grant number LY13G030002], Philosophy and Social Science Project of Zhejiang [grant number 11JCLJ01YB] and Qianjiang Talent Plan [grant number QJC1402004]. Yiyun Wu (corresponding author) acknowledges the financial support from the Natural Science Foundation of Zhejiang [grant number LQ14G030011]. Xiwei Zhu acknowledges the financial support from National Natural Science Foundation of China [grant number 71373235] and Qianjiang Talent Plan [grant number QJC1302007].

Notes on contributors

Deming Luo

Deming Luo is an associate professor in the College of Economics, Zhejiang University. He has worked on a wide variety of topics in economic theory, corporate finance, and applied economics.

Yanjun Liu

Yanjun Liu is a PhD student in the College of Economics, Zhejiang University. Her research interests focus on trade and economic geography.

Yiyun Wu

Yiyun Wu is an assistant professor in the Social Sciences Academy, Zhejiang University. She works on policy oriented issues in development economics and China economy.

Xiwei Zhu

Xiwei Zhu is a professor in the College of Economics, Zhejiang University. His main research areas include new economic geography and China economy. He has published in international journals, such as Japanese Economic Review, Papers in Regional Science, and Regional Science and Urban Economics.

Xiangrong Jin

Xiangrong Jin is a professor in the College of Economics, Zhejiang University. He has published widely on economic theory and development with particular focus on industrial policy and regional development in China.