LAUSR

In the late 1990s, China’s Yangtze and Yellow River Basins suffered devastating natural disasters widely attributed to the degradation of soil and water resources. The Government of China responded with a number of major environmental programs, the most expensive and influential of which, the Grain for Green (GfG) Program, was implemented widely from 1999. Under the GfG Program—also known as the Sloping Land Conversion or Conversion of Cropland to Forest Program—the central government compensates farmers to convert cropland on steep slopes or otherwise ecologically sensitive areas to forest or grassland. Its long-term success depends on households’ ability to make sustainable changes to their household income streams and income diversification strategies. In this paper, we use a difference-in-difference estimation approach to examine the role of migration as a household-level response to the GfG Program, testing the extent to which individuals migrate following a reduction in land available for farming. Importantly, we exploit 15 years of data on migration decisions and establish that participating and non-participating households were on parallel migration paths before the program, thus refuting a key threat to causality in a difference-in-difference model. We find that participating families do, in fact, choose migration as an income diversification strategy more frequently than non-participants. The program effects varied over time but peaked post-Great Recession in 2011 when migration rates in GfG households exceeded those of non-GfG households by 5.9% points (p = 0.003) or about 26%. Our findings should encourage policymakers that families are making long-term adjustments to their livelihood strategies to avoid poverty in anticipation of the eventual withdrawal of government supports.