Does Expressway Consume More Land of the Agricultural Production Base of Shandong Province?

The effect of expressways on cultivated land is ambiguous. Many studies conclude that building and upgrading expressways increases pressure on cultivated land while others find expressways reduce the rate of cultivated land loss. In this paper, we use satellite remote sensing images of cultivated land in Shandong province of China to test whether the existence of expressway in 2005 affected the level of cultivated land in 2010 and the rate of change from 2005 to 2010. To account for expressway access for each of our 1 $$\hbox {km}^2$$ km 2 (‘pixel’) units of cultivated land we measure whether or not and what type of roads penetrate the ‘watershed’ in which the pixel lies. These watersheds allow more plausible measures of accessibility than those traditional ‘crowfly’ distance measures that ignore topography. To account for possible confounding we also use 24 additional covariates. Although simple univariate OLS regressions analysis show that cultivated land is always lower while cultivated land increasing rates are higher either when there is an expressway, these results are not robust. Controlling for all of the covariates and also using recently developed covariate matching techniques to estimate treatment effects, we find that expressway can most safely be described as putting a positive impact on cultivated land changes.