Do land markets improve land-use efficiency? evidence from Jiangsu, China

ABSTRACT

Inefficient use of scarce and fragmented land challenges the sustainability of agriculture. Land markets may improve land-use efficiency. In recent years, China has employed various instruments to promote land markets. This paper investigates whether land markets affect households’ land-use efficiency, based on data from 1,202 farm households in Jiangsu Province. The measure of land-use efficiency was derived from a stochastic frontier production function, and a control function approach was employed to correct for selection bias. The results indicated that many households are using land inefficiently. While renting in land increases land-use efficiency, it is not affected by renting out land, implying that households are not giving up land for efficiency gains. We also provide suggestive evidence that the positive effect of renting in land results from abundant agricultural labour due to labour market failure.

KEYWORDS:

• Land market
• technical efficiency
• stochastic frontier analysis
• control function

JEL CLASSIFICATION:

• Q12
• D24

Disclosure statement

No potential conflict of interest was reported by the authors.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

¹ Some authors have employed a linear probability model in the first step of the control function for dichotomous variables (e.g. Brasselle, Gaspart, and Platteau (2002) and Rao et al. (2017)). However, Lewbel, Dong, and Yang (2012) point out that a linear probability model in the first step could lead to biased estimates if the outcome variable in the second step is binary or limited. Brasselle, Gaspart, and Platteau (2002) also point out that key assumptions (e.g. homoscedasticity) of control functions would be violated if the first step equation were a linear probability model for non-continuous variables..

² One hectare is equal to 15 mu.

³ Due to data limitation, we were unable to measure these inputs in physical units. We followed Ma et al. (2014) in their use of monetary units.

⁴ With the ideal situation as the base, it is calculated as follows: $\frac{Z_i-Z_i^M}{Z_i^M}=\frac{Z_i}{Z_i^M}-1=\frac{1}{LU{E}_i}-1=\frac{1}{0.930}-1=7.53\mathrm{\%}$.

⁵ Since the determinants of household participation in land markets are beyond the interest of this paper, the first stage estimates of the control function approach are reported only in the appendix in Table A1.

⁶ We performed the test based on the results of the Cobb–Douglas and translog production functions with (non-robust) standard errors.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China (grant numbers 71673144 and 71773046), the Specialized Research Fund for the National Social Science Fund of China (grant number 18VSJ060), Humanity and Social Science Youth Foundation of Ministry of Education of China (grant number 18YJC790110), China Postdoctoral Science Foundation (grant number 2017M620217) and the 111 Project (B17024).