The Effects of Untying International Food Assistance: The Case of Canada

Most donor countries historically linked international food assistance programs to their domestic agricultural support policies through tying policies that required donated food to be purchased in donor‐country markets. Tying policies have been shown to reduce the efficiency and effectiveness of food assistance along a number of dimensions, including cost, timeliness, and cultural appropriateness. We investigate how the Canadian policy to untie food assistance affected Canadian shipments. We use an operation‐level dataset from the World Food Programme to investigate how procurement sources and commodity compositions changed after untying, and we build an empirical model of Canadian food assistance shipments that is used to compare observed shipments to counterfactually‐tied shipments. This comparison reveals positive effects of the untying policy at the extensive margin (number of recipient countries) and no significant effects at the intensive margin (quantity of food) when estimated over our entire sample. We observe positive effects of untying at the intensive margin (quantity of food) when comparing shipments during the untied regime to shipments in years directly preceding the policy change. Our results show that Canada delivered more emergency food assistance, more frequently, after untying. The results of the Canadian experience inform ongoing debates about untying food assistance in the United States and untying other forms of foreign aid.