Abstract
Beginning in the early 2010s, many of the cities in China dominated by state-owned enterprises began to attract investment from the country’s more developed east coast, thereby opening the way for their economic development. However, this transition was not consistent across all regions. How can regional variation in this transformation be explained? This article argues that the capacity for local governments to make relatively autonomous decisions regarding the state-owned economic sector has been essential for economic transformation. This is because it is only when they have such independence that local officials will be able to fully employ the policy tools necessary for the institutional reforms that allow them to concentrate the resources needed for development. To demonstrate this argument the article uses a controlled comparison of development in Mianyang and Deyang, two neighbouring cities in Sichuan Province.
Acknowledgements
Previous versions of this manuscript have received comments from many colleagues. The authors would like to thank Jingyang Huang, Chengyuan Ji, Wei Chen, Lingling Qi, Qing Wang, anonymous reviewers, and the journal editor for constructive feedback.
Disclosure Statement
No potential conflict of interest was reported by the author(s).
Notes
1 The concept of controlled comparison was introduced by George and Bennett (Citation2005). According to them, the controlled comparison is a case-based research method for causal inference. The typical controlled comparison designs are method of agreement and method of difference. In this study, since it is using two neighbouring highly similar cities to test the importance of state autonomy, it is using method of difference.
2 Frequencies were calculated from the Mianyang Yearbook from 2006 to 2019 and using NVivo 12.0 Plus (Mianyang Yearbook Compilation Committee Citation2006–2019). The Yearbook includes five sections: photos, special editions, annual review of government agencies and relevant institutions, chronicle of events, and appendices. Because the annual review of government agencies and institutions summarises the priorities and achievements of the year’s work, it was used here to calculate the frequency and coverage rates. From the fieldwork, it is known that the official in charge of drafting the annual work report usually bases it on the previous year’s report. This informal rule of work leads to a high degree of comparability in the structure and length of annual reports. As a result, it is reasonable to take changes in word frequency across these highly similar texts as an indication of the distribution of government attention. Coverage rates were calculated by dividing the frequency of a term’s appearance by the total number of sentences in the text.