Poverty and Land Redistribution: Quasi-Experimental Evidence from South Africa's LRAD program

We estimate the impact on consumption of South Africa's Land Redistribution for Agricultural Development (LRAD) program, which provides capital and other forms of assistance to beneficiaries to enable market-assisted transfers of property rights from large landowners to the rural poor. Counterfactual outcomes are derived by screening out applicants with low probability of admission into the program, and then non-parametrically matching beneficiaries to applicants that are still in the pipeline to receive assistance. We find that land transfers associated with LRAD lead to very strong benefits for program participants. Accounting for heterogeneity in the length of exposure to the program, we estimate treatment effects of the program that peak at approximately 275 Rand per capita monthly consumption. Assuming a discount rate of 5%, this estimate translates to a discounted gain in monthly per capita consumption of about 50% after three years of exposure to the program.