Does where you are from affect how you land? Evidence from land transactions of Chinese manufacturers

ABSTRACT

Leveraging the unique matched land transaction and firm survey data, we empirically examine if firm ownership will affect the access to land factor for Chinese manufacturers. The state-owned firms display no advantage in paying less when acquiring land, while it is more costly for foreign-owned firms. Consistent with this finding, we find land acquisitions for foreign-owned firms are more likely to take place via the nonmarket-based transaction in the early stage of China’s land market reform with the goal of eliminating corruption. The research suggests that potential land miss-allocation (if any) could be resulted from ‘tax’ to the foreign firm rather than ‘subsidy’ to state-owned firms.

KEYWORDS:

• Land prices
• price distortion
• firm ownership
• Chinese manufacturers

JEL CLASSFICATION:

• E22
• R14
• R52

Acknowledgments

The views expressed herein are those of the author and do not necessarily reflect the views of the State Administration of Foreign Exchange of the People’s Republic of China. All errors are mine.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

¹ The summary statistics is provided in Table 1.

² We consider the equity from Hong Kong, Macau, Taiwan, along with other foreign countries as foreign equity.

³ We use firm location and firm name provided in both datasets in the matching process.

⁴ The number of matched firms accounts for 12.55% and 26.7% of the total number of listed firms purchasing lands in 2007 and 2013.

⁵ There are only a few matched firms with multiple transactions, so we cannot control firm fixed effect. We are still working on matching for other years and matching firms across years, which takes tremendous time.

⁶ It suggests that foreign firms are likely to be treated unfairly by charging a higher land price compared to firms of other ownership.

⁷ Due to lack of knowledge of the regression coefficients’ distribution, we statistically examine the coefficient differences through bootstrapping by 1000 times and obtain the regression coefficients sample in year 2007 and 2013. We then compare the mean differences between the two sample to find out whether the average mean of the two sample with obtained coefficients are the same or not.

⁸ There is also other paper using the land transaction data in China. For example, Chen et al. (2016) observe that firms invest more in commercial land instead of industrial land due to the real estate shocks which could lead to potential capital misallocation. We also find similar patterns from our matched land transaction and manufacturing survey data. For example, in Table 2, the value share of land transactions for Commercial Real Estate has greatly risen from 2.88% to 14.8% due to the housing boom.