Urban food waste in the U.S. is almost exclusively bound for landfills, with significant environmental and the economic consequences. Contrary to this linear flow, cities across the developed world are implementing organics collection programs (OCP) that transform waste into a resource. These systems convert excess food into inputs (e.g. biofuel or soil amendment) for other processes (i.e. food and energy production, water reclamation) thereby approximating naturally occurring closed-loop nutrient systems. However, efforts to address the flow of food waste into landfills have largely focused on technical solutions that alter the waste management systems of provision, largely ignoring the social aspects of food waste generation and source separation. Drawing on the Theory of Planned Behavior and concepts from social marketing research, results from longitudinal repeated-measures experiments indicate that providing supportive infrastructure is necessary for residents to act on their "intention" to divert food waste from landfills. The experiments are also the first to confirm that communicating the new community norms of separation in the context of a new OCP will increase separation behavior. In sum, closing the loop on urban food systems demands transformations of the systems of provision as well as the social aspects of waste systems. This research traces the evolution of organic waste policy on two continents using problem-framing as a lens through which to understand how different characterizations of food waste as a problem (i.e. as one of disposal efficiency, risk management or value recovery) result in different policy responses with varying degrees of linearity in urban food material flows. This dissertation contends that successfully closing the loop on urban food systems will demand new problem-framing that employs systems-thinking at the highest levels of policy-making, transformations of both the systems of provision and social practices, and an understanding of the relationship between social and natural systems (i.e. socio-ecological relationships).