Under what conditions do host governments enable or obstruct the delivery of goods and services to people in need in the aftermath of humanitarian emergencies? Every year, donors give hundreds of millions of dollars to United Nations’ agencies and international NGOs to deliver humanitarian aid to people in need. These humanitarian organizations must gain the permission of the host government to reach their intended beneficiaries, but neither practitioners nor scholars account for the role host governments play in enabling or obstructing humanitarian organizations’ operations. While agents of the state are sometimes quick to collaborate with international actors, in other circumstances they deny emergencies exist and impose restrictions that prevent humanitarians from delivering aid. By tracing the process through which governments decide to acknowledge an emergency exists, allow international organizations to provide aid, and impose restrictions on aid after it arrives within their borders, I show that host governments shape whether and how humanitarian aid is distributed, the time it takes organizations to provide aid, and who is able to access the benefits of aid.

My theory explains that states’ decisions are driven by leaders’ need to maintain a reputation for competence among both domestic constituents and international donors. States collaborate with humanitarian organizations when doing so will improve their reputation for competence and deny the existence of humanitarian emergencies when leaders fear donors or constituents will blame them for the existence of the emergency. Collaborating with humanitarian organizations strengthens governments’ reputation for competence in response to fast-onset disasters but undermines governments’ reputation for competence in response to slow-onset disasters. To avoid damage to their reputation for competence, host governments impose restrictions on humanitarian organizations that prevent these organizations from revealing novel, credible information that would damage the government’s reputation for competence if revealed.I leverage cross-national data, original surveys, and in-depth interviews to assess the theory and its implications. I first analyze cross-national data to test whether governments are more likely to obstruct the delivery of humanitarian assistance in response to slow-onset emergencies compared to fast-onset emergencies. I then use interviews with representatives of donor countries and an original survey of 530 humanitarian professionals to describe the various strategies governments use to restrict humanitarian organizations and illustrate the consequences of these restrictions for humanitarian organizations’ operations. Drawing nine months of fieldwork, I use the case of Niger to illustrate the logic underpinning government decisions to restrict or collaborate with humanitarian organizations. I use a survey of over 400 Nigerien government officials to test my expectations regarding host-government officials’ attitudes and behaviors toward humanitarian aid.