Abstract

A recent paper by Wang et al. (2021) argues that management intensive grazing (MIG) practice adoption might be ‘ … a key factor for restoring marginal croplands to permanent grassland cover … ’ in the United States Northern Great Plains. The matter is important for land use policy because U.S. Federal and State governments actively seek to promote grass cover through a variety of policy instruments. In this note we show that the significant positive coefficient on MIG adoption they estimate in a multivariate ordered probit model does not indicate a valid causal relationship. Their estimates are vulnerable to bias and inconsistency due to potential endogeneity and so do not support their policy inference.